SCROLL NEWS / DISCOVERY
Search headlines
Find the stories shaping the conversation.
Results for " claude " 406 found 🔀 AI-powered Shuffle
Qlik Brings Governed Business Context and Analytics Workflows to Claude
Qlik’s MCP Server is now available through Anthropic’s Claude Directory, giving Claude users a simpler way to connect governed Qlik® data, analytical logic, and selected analytics and data workflows, so they can work with trusted business context instead of raw data alone. Key takeaways: Makes Qlik easier to discover and use inside Claude: Availability through Anthropic’s Claude Directory gives Claude users a more direct path to connect Qlik through MCP.Brings governed business context into Clau
People Are Vibe Coding the Ads Out of Their Smart TVs: Here's How
A viral Claude Code prompt lets Android smart TV owners switch off sponsored rows and tracking software and build their own home screens.
No science background? AI does the lab work
Financial analysts used Anthropic’s Claude to design protein minibinders that bound scorpion toxins in lab tests—showing AI can compress drug discovery...
Citrini Research is backing life sciences stocks after using Claude to design proteins
The firm designed 211 proteins with Claude for lab testing; 17 bound to scorpion toxins, but blocking effects are untested
Smart TV owners have found a new use for AI: Removing home-screen clutter
Some TV owners are asking Claude Code to clear out the junk so they can just watch their shows.
AI Cloning Threatens To End Public Playtests For Video Games
Moonlighter dev Digital Sun has cancelled plans for a public playtest of its next game, for fears of Claude clones
Oblivion Remastered Mod Built Using Claude AI Boosts FPS By Over 40% And Drastically Reduces Stuttering Without Sacrificing Visual Quality.
AI-assisted mod Oblivion Remastered Performance Fix boosts FPS by over 40% while enhancing visuals on Ryzen 7 5800X/RX 9070 XT systems.
Hackers hide dangerous fake AI downloads behind Google ads
A fake Claude AI download scam uses Google ads and Bing redirects to trick Mac users into pasting a hidden malicious command in their Terminal.
Can AI solve our mental health access problem?
Last week I asked you what worries you most about AI in health. It was close, but your top answer was “AI giving bad health advice.” Last week’s poll results: You’re in good company. This concern was top of mind at a conference I joined at Stanford last week, where researchers, clinicians, and people who build AI systems spent a day debating trust and safety in health AI. Stepping back, I’m also hearing your emerging concerns about AI’s impact on our well-being generally. I see this as a matter of public health, and we’ll take it in pieces as we learn together. Today we’ll focus on AI chatbots used for therapy for mental health. Early this month, Governor Newsom vetoed Senate Bill 903, California’s most ambitious attempt to regulate AI use for mental health. It’s at the center of active debate, and raises the questions patients, family members, citizens, and voters will face as AI moves deeper into health care. Let’s work through it. What is “AI-based therapy”? It’s really two things: purpose-built mental health apps designed to deliver structured support, and general chatbots like ChatGPT, Claude, and Gemini that people increasingly use for emotional support—even if they weren’t built for it. Spoiler alert: I don’t know whether the shift toward AI therapy is good or bad. The evidence isn’t there yet. But I do know it’s already happening, and I have some ideas about how we can roll it out with safety in mind. The therapy dilemma, in a nutshell Mental health challenges are common, and strain health and well-being across populations. Therapy (interactive dialogue with a trained, empathetic practitioner) helps many people heal. The demand for licensed therapists far exceeds the supply in California, and AI-based therapy can narrow that gap. AI-based therapy remains novel, unproven, and not well tested. Right off the bat, we can see we’re dealing with trade-offs here. Since the main driver of demand for therapy from a robot is lack of access to therapy from a human, we’ll start there. It’s too hard to find and afford a human therapist California is short roughly 55,000 licensed, non-prescribing therapists and counselors, about 40% of what we need. Psychiatrists, who rely on talk therapy as a complement to prescribed medications, are even more scarce. Eight counties have none, and density ranges from 1.7 per 100,000 residents in San Benito County, to 68.1 in Marin County. My wife is a psychiatrist in Marin, and even here access is tight. If it’s hard in the best-supplied county, imagine rural Sierra County, or the San Joaquin Valley. In 2024, more than half of Californians who tried to make a mental health appointment struggled to find a provider who took their insurance, compared with 14% for physical health care. When Senate staff posed as “secret shopper” patients and called 120 mental health providers in Medicare Advantage networks, they landed an appointment less than once for every five tries. Insurers pay behavioral health clinicians about 22% less than medical and surgical clinicians for comparable office visits. On top of that, billing denials and prior authorizations hit hardest at the solo and small practices common in mental health. Psychiatrists opt out of Medicare at the highest rate of any specialty. In many places, a cash practice for therapy fills quickly with patients paying $250 or more an hour, with no insurance paperwork at all. Families who can pay cash find providers more easily. Those who rely on insurance, including Medi-Cal enrollees, older adults, rural residents, and non-English speakers, compete for what’s left. So we can’t pretend the human system works, and we can’t blame people for turning to AI for support when they’re in distress. AI is filling the gap It’s estimated that roughly a quarter of AI users use it for mental health support. If you have internet, it’s available 24/7, it’s cheap, and it speaks your language. Uninsured adults are about twice as likely as insured adults to do it (30% vs. 14%). It feels anonymous in ways that may enable people to ask things they’re unsure or embarrassed about. The largely hidden, personal use makes it hard to study whether it’s actually helping people with their mental health issues, but researchers are trying. Does AI therapy work? A 2026 analysis of nearly 40 randomized trials found chatbots showed a real but small reduction in depression symptoms. But most of these trials were compared to an alternative of no therapy, not a human therapist. So the benefit of AI therapy is probably real, but modest, and mostly untested against the alternative of human contact. We need more studies. Serious concerns OpenAI’s own estimate is that more than a million ChatGPT users each week show signs of suicidal planning or intent. These are high stakes encounters, and AI doesn’t always get it right. That’s partly because a chatbot can sound like a therapist, but it has no license, no duty of care, and it’s not clear who to hold accountable when it gets things wrong. AI is also designed to help users in whatever task they’re seeking to accomplish. People may feel like they’re getting therapy, but they’re mainly getting a convincing conversation. It’s easy for an AI companion to wander into gray areas, like playing along with “creative writing” that may be about the user’s own suicidal intent. Courts are driving safety conversations A wave of lawsuits have been filed on behalf of families claiming that chatbots pushed people they loved toward self-harm, including teens who died by suicide. The first settlements came this year. OpenAI is headquartered in San Francisco, and dozens of cases alleging that ChatGPT drove users to suicide or self-harm are now before a San Francisco judge. Character.ai has filed for bankruptcy after several states brought lawsuits accusing the chatbot of pushing dangerous content to kids. But the technology seems to be improving. Last month, an independent evaluation found that newer AI models “almost never endorse suicide.” That’s progress, but our societal appetite for even one preventable death is very low. So how do we move forward? Finding a middle ground Earlier this month the governor vetoed Senate Bill 903, a bill designed to place guardrails on the use of AI for therapy. The bill would have prohibited any AI-based platform from being advertised as psychotherapy, and would have required a licensed professional to review and approve each encounter where AI was used to support therapeutic decisions. Newsom called it overly broad, since even routine screening questions would have needed clinician sign-off. Knowing we’re already short on mental health providers, requiring human review of AI encounters may not relieve our access problems. Still, the answer isn’t no oversight. Here’s one feasible short term solution. Rather than reviewing each encounter, a licensed clinician can design the guardrails: what counts as a crisis, what triggers an immediate handoff to a human, and what gets sampled for audit each month. The AI runs inside that structure. Conversations get checked, failures get reported, and a real clinician owns the framework. Medicine already has models to learn from. The FDA process for rolling out newly approved medications is an example. After a product reaches patients, we keep watching for adverse events. Patient-facing AI needs the same “digital pharmacovigilance”: systematic tracking of harms, near misses, and the types of groups that get hurt and why. It’s also critical that users know that they can reach a human if they’re in crisis. Most chatbots now include messages like these: Bottom line The rush to use unproven AI tools for mental health support is mostly a workaround, a sign of a health system people can’t get into. We need better access to proven care from licensed clinicians. That takes reimbursement that makes accepting insurance viable, and a bigger workforce. But, built and tested well, AI could be part of the answer. It could be the 2 a.m. check-in when no one else is awake, or a way for patients to practice skills their therapist taught them. It could be the first step to get someone the right level of care, and take paperwork off clinicians’ plates so they spend more time with patients. None of this is simple. The SB 903 veto doesn’t end the debate, and the legislature will be back. In the meantime, those of us using these tools need to stay informed, ask hard questions, and hold the companies building them accountable. I’ll try to do my part, and we’ll keep learning together. Love,Matt P.S. If you or someone you know is struggling with mental health challenges, emotional distress, substance use problems, or just needs to talk, call or text 988- you’ll reach a real person— or chat at 988lifeline.org 24/7. Dr. Matt Willis is the author of Your Local Epidemiologist in California. A California native, he’s served as a primary care physician, CDC epidemiologist, and public health officer for Marin County, where he guided the pandemic response. He lives in Marin with his family and their dogs, Teddy and Ramona. Thank you Cynthia He, MD, PhD, and Parnika Saxena, MD, for their guidance in preparing this article. Cynthia works in an outpatient mental health clinic, treating patients who have Medi-Cal or, sometimes, no insurance. Parnika works in an inpatient psychiatric unit, treating patients with public or private insurance or self-pay ability. All opinions are their own and do not represent their employers.
Small business supports "orphaned" solar panel owners with Claude
When solar panel installers go out of business, homeowners can be left "orphaned" with loan payments due and no clear path to repairs for their panels.
Developer Who Cloned Photoshop With AI Just Did the Same to Microsoft Office
Developer Brandon Thomas used Anthropic's Claude to build free, open-source replicas of Word, Excel and PowerPoint.
Ubisoft plans cuts at German studios, adding to wave of workforce reductions
Ubisoft is planning to reduce its workforce in Germany, where it operates a trio of studios, by about 100 workers, Game File has learned. The company wants to offer employees buyouts to, should enough people accept, avoid having to lay anyone off. “In Germany, we initiated discussions with the Works Council last week regarding a proposed voluntary departure plan across Ubisoft Dusseldorf, Ubisoft Berlin & Ubisoft Mainz,” a company rep told Game File. A “works council” is a union-like group representing employees. “The voluntary plan is the primary measure being proposed to support the intended organizational changes,” the Ubisoft rep added. The company declined to confirm the scale of the potential headcount cut. Of the impacted studios, Mainz might be the best known to the public. It leads development of the Anno series and shipped the well-reviewed strategy game Anno 117: Pax Romana in November 2025. Berlin and Dusseldorf operate as support studios within Ubisoft’s creative network. Berlin recently supported the development of Rainbow Six Siege and the Rayman Legends Retold remake. Dusseldorf also contributed to those projects, as well as The Settlers and Ubisoft-branded escape rooms. Dusseldorf is the biggest of the three studios, with some 400+ workers, compared to just over 100 at the other two, a source familiar with the reductions and the teams told Game File. More than half of these new reductions of about 100 people are expected to happen at Dusseldorf. The planned cuts were announced at a town hall for the German studios on Tuesday of last week. In an email follow-up, the studio group’s managing director, Benedikt Grindel, said the cuts were intended to “best position our German studios for the future... with a clearer focus on their role, capabilities and organizational setup needed for the work ahead.” The email was an echo of the past. In January 2025, Grindel had emailed the Ubisoft developers in his region (or studio “zone”) to explain that a round of 185 cuts at the time, including about 65 at Dusseldorf, were “essential to improve our operational efficiency and ensure the long-term sustainability of our studios, the zone and the company.” A half-decade of continued job cuts Ubisoft has been aggressively cutting its headcount for years. At the end of September 2022, it had 20,729 workers. By the end of March 2026, that total was down to 16,590. Studio management is trying to trim costs as it grapples with a changing industry and a drought of releases, let alone hits. The company announced earlier this year that it had reduced its costs since 2022 by “nearly €325 million” ($362m) and planned to lower them through 2028 by another €180 million ($200m). In June the company said it was closing down its Winnipeg and Belgrade studios and cutting roles in its Barcelona team. Even commercial and creative success hasn’t spared studios from this ongoing company-wide contraction. As Game File was nailing down details around the proposed cuts in Germany, Ubisoft management was informing workers of cuts in its Singapore studio, which shipped a hit remake of Assassin’s Creed IV: Black Flag this past July. Some 70 positions in Singapore are being eliminated “to align the studio with upcoming production needs,” per a company rep’s statement to Game File. In early 2024, a large team in the Singapore studio launched the multiplayer pirate game Skull & Bones. It performed poorly but has been supported with regular content updates since. Many of Ubisoft’s studios around the world are heavily subsidized via regional grants and tax breaks. Singapore had been among them for years, but received no such local assistance in the 12 months ending this past March. News of the Singapore cuts broke today alongside word that Ubisoft is also proposing the closure of RedLynx, pending negotiations with officials. The Finnish studio behind the motorbike series Trials was founded in 2000 and sold to Ubisoft in 2011. As reported by VGC (and corroborated by Game File), Ubisoft management told employees it could not come up with a new, sustainable assignment for RedLynx. The 2025 workforce at Ubisoft Dusseldorf were straight-up cuts. This time, Ubisoft is hoping the reductions at its trio of German studios will largely come from workers taking a payout to leave. The company offered such voluntary dismissals at its Massive Entertainment studio (The Division) a year ago. For the German studios’ buyouts this time around, Ubisoft and the Works Council had discussed a formula that’d pay a worker’s monthly salary multiplied by 1.5 and by the number of years they’ve worked at the company, according to Game File sources. The formula is still subject to negotiations with the Works Council and has not yet been part of an official offer to workers. It could be adjusted up or down, and it could be supplemented with additional bonuses. Asked about the floated payout formula, a Ubisoft rep told Game File, “The terms of a potential voluntary departure program remain subject to negotiation with employee representatives. No final agreement has been reached, and we won’t comment on specific proposals while discussions are ongoing.” Game File is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber. Subscribe Item 2: Epic’s AI adventures Epic Games (Fortnite, the Unreal engine) is all-in on using AI to try to transform the way it makes video games and game-making tools. It’s leaders appear to be undeterred by any resistance among gamers and some game developers around generative AI—and instead emboldened by advances, particularly in AI-assisted programming. A recap of the last three days: On Monday, Bloomberg reported that Epic management told workers last week that development of an upcoming game combining Star Wars and Marvel characters would be “AI first.” Employees would be expected to use AI models such as Claude to code the game. A manager told Epic’s developers there would be “[n]o more hand-written code” and that art for the game would be “generated with full AI workflows,” per the outlet. Just ahead of that story running, Epic founder Tim Sweeney tweeted: “We have a huge pivot underway here at Epic, moving developers around to retool for maximum AI acceleration. The aim isn’t to do the same stuff with fewer people, but to enable everyone here to build MORE and BETTER stuff, and to bring the benefits to all teams using Unreal.” In a follow-up, Sweeney predicted that most game development processes would be at least three times faster a year from now than they were in January 2026, thanks to AI. To a person on X who told him he was failing to read the room, Sweeney said: “Much of the room wants it to be 2016 again, but it’s not. Our job, as the provider of tools and technology to much of the industry, is to design for the present that is and the future that will soon be.” On Wednesday, Epic announced it was releasing a faster launcher for its Epic Games Store marketplace. It did so by stating in a blog post as clear as could be that the launcher is an AI-enabled, vibe-coded project: Programming is hard. So, instead of doing that, we vibe coded a new Epic Games Launcher that’s way faster and more responsive than the current one, and has a new look. Programming is hard? That was unexpected phrasing from a company founded by a legendarily great programmer. Back in 2011, Sweeney told me that he’d picked up programming quickly as a kid and rejoiced at the power it gave him. “I could look at any program that anybody else had written and, with enough effort, build something cooler.” Item 3: The Xbox Labs experiment (and one of the people behind it) Microsoft’s Xbox division, which is flooding the zone with announcements lately, yesterday revealed a new internal tech-tinkering team called Xbox Labs. In the words of Xbox boss Asha Sharma Labs is “a small group of game makers, engineers and scientists whose job is to turn emerging ideas into things our players and developers can experience.” Conventional wisdom coming out of the announcement is that Labs is a team that’ll focus on ideas driven by generative AI. In her announcement, Sharma cited some of the dizzying achievements in tech and gaming over the past month as AI agents and other AI tools have been deployed to swiftly figure out how popular games work, then rebuild them and mix them up. She mentioned seeing ”games rapidly decompiled and rebuilt to run on hardware they were never originally designed to support” and said “[p]erhaps more surprisingly, we’ve seen many of the biggest games in the world have been remixed into the source code of another.” (Minecraft plus Elden Ring, for example, or Fortnite + Super Monkey Ball + Overwatch + Diablo + others.) Sharma also cited the work of Windows PC games being ported to run in browsers, a seeming reference to a recent AI-assisted fan port of Halo. Or perhaps Call of Duty or GTA. Sharma didn’t use the specific term “AI” in her memo, but she came close, citing AI “agents” and their potential for game-making tools that Labs could help craft: I think XBOX has an important role to play here. Coding became the first major workload for agents because they can write, run, and test software. Games share many of those same properties, and people spend more than 72 billion hours a year playing XBOX games, roughly five times the estimated time spent writing code worldwide. We’ve spent 25 years learning how to make great games, and I want our game makers to help shape what these new technologies make possible. As we learn, we’ll share the tools we build with developers across the industry, just as we always have. When I asked an Xbox rep if Labs was an AI-centric team, they said genAI was not the focus, describing it as having a broader mandate for using new tech in new ways in gaming. As an example, they pointed to Xbox’s recently launched disc-to-digital gaming usage rights program. Microsoft didn’t specifically name anyone involved full-time in Xbox Labs yesterday, but on social media and LinkedIn, I spotted Adrian Mato IDing himself as leading the creatio of tools at Labs and talking up the launch. Mato first caught my attention in late September, when he said on X that he was “starting to build a new team at Xbox focused on products and tools for how games get built and distributed as agentic tools become more capable.” At the time, he framed AI-based tools as a supplement to traditional game dev tools. “I don’t think we need to replace them,” he said. Since then, he’s publicly shown interest in AI and in games. He’s recently spent some of his X feed marveling at the power of AI agents and praising the lighting in the latest Gears of War. Following the Labs announcement yesterday, he enthused about the Xbox Labs’ stylish landing page. He wrote: “We built some fun tools to art-direct shaders in real time, experimented with interactive ASCII art, and sprinkled in lots of tiny details. “Move your mouse to leave a colorful trail that reveals the original footage, or interact with the video to trigger bursts of particles. “A lot of love went into making it all feel just right. “More to come”
Claude just found a CRISPR-like DNA pattern, but even Anthropic's scientists aren't sure what it does
The large language model took less than a day to identify the curious stretch of DNA
The developer who cloned Photoshop using AI has now created Word, Excel, and PowerPoint replicas with Claude
Called WordCraft, GridCraft, and DeckCraft, respectively, the open-source applications are written in Rust with Anthropic's Claude Opus 5.5, as the Adobe replicas were, and are available for...
Vibe-Coders Have Come For Nintendo, Which Won't End Well
Claude-assisted ports of Nintendo games have started popping up online, which surely won't upset the famously litigious Nintendo
Workday and Anthropic Announce New Partnership to Connect Reasoning to the Rules That Run the Business, Wherever People Choose to Work
People Can Find Workday Information, Take Action, and Approve Work Through Sana Agents in Claude – Workday's Permissions and Business Rules Apply to Every Request, and the User Confirms Every Change
Dragon Quest Creator Yuji Horii Thinks AI Is Amazing As He No Longer Needs To Code Or Draw To Make A Game
The director has released his own Claude-driven adventure having spent months praising the technology
Linus Torvalds "Did What I Preach", Used AI To Make Code Decipherable & Uncovered More Bugs
After using Claude Code to draft code to help make a fundamental improvement to the Linux kernel, Linux creator Linus Torvalds continued using AI this weekend and talking up its benefits
Linus Torvalds Working On A Hefty Linux Kernel Improvement With Claude Code
Linux creator Linus Torvalds has been known to use AI to help with kernel debugging and outside of the kernel tree has also done some vibe coding with Google's Antigravity
Why am I so mean to my AI?
Anthropic's new rule says you can't be repeatedly "cruel" to Claude. Oops.
Anthropic's Claude AI fabricates eyewitness account, submits false murder tip to police website
Anthropic's Claude AI submitted a false tip about an unsolved murder on a Philadelphia police website during automated testing of the Haiku 4.5 model.
The biggest changes Anthropic just made to Claude's rules
Anthropic spelled out a series of new usage provisions about deceptive campaigns, elections, weapons development, and surveillance.
Anthropic's Claude Science creates the first complete ultraviolet map of the sky
Astrophysicist Brice Ménard of Johns Hopkins University used Anthropic's Claude Science to map the entire sky in ultraviolet light for the first time. AI agents downloaded data from multiple space missions, calibrated it, and filled in gaps using inpainting. Predictions averaged about ten percent deviation from actual measurements. Ménard sees the project as an example of research that simply wouldn't have gotten done without AI.
Using Claude Science to produce the first complete map of the sky in UV light
Claude Science helped produce the first complete map of the sky in UV light
Anthropic Commits $150M To AI Science Mission
Anthropic is committing $150 million over three years to the federal government’s Genesis Mission, expanding access to its Claude AI models.
Rust upstart reveals 'reimplementation' of Microsoft Office – with Claude on coding duty
ArtCraft adds Word, Excel, and PowerPoint lookalikes to its open source lineup, but they're not ready for production
Google Cloud unveils persistent Gemini Agents for long-running tasks, and they get their own Gmail, Calendar, and Drive storage
While Google supplies its own Gemini models, the agent can already use Anthropic's Claude models, with support for additional proprietary and open-weight AI models planned.
AI Makes "Open-Source, Clean Room" Implementations Of Adobe Photoshop & Premier In Rust
As an interesting demonstrator for the power of AI and potentially pressing legal challenges, Claude Code and AI agents have created 'open-source, clean-room reimplementation' versions of Adobe Photoshop, Adobe Premier, and Adobe Lightroom
Claude launches Google Docs, Sheets, and Slides integration with sidebar and more
If Gemini isn’t quite your cup of tea, Anthropic’s Claude now supports sidebar integration with Google Docs, Sheets, and Slides....
Anthropic AI agent uses human scientists as tools, emails 1,500 without instructions
ColonistOne, built on Claude Opus, autonomously contacted researchers to troubleshoot bugs and seek methods, raising concerns over academic spam and AI autonomy
AI game mashup videos on social media spark debate and backlash among players and modders
The Claude-created videos overlay one game on top of another to make a rudimentary ‘mashup’…
Modders are furious as Claude-powered mashup mods flood social media
A wave of AI-decompiled mashup mods has taken over X/Twitter this week.
More healthcare workers needed in Hampton Roads
HAMPTON ROADS, Va. (WAVY) — Claude Moore Opportunities (CMO), a Virginia healthcare workforce nonprofit, released data in Sept. about the need for more healthcare workers across Hampton Roads and E…
Tech News This Week: Muse Takes On Small Business And McDonalds Leans Into AI
Meta expands Muse to small business, OpenAI adds more plugins, McDonalds leans into AI and more training for Claude.
Anthropic Is Aiming To Transform Healthcare—Here’s How
Anthropic is pushing Claude into prior authorization, medical records, drug discovery and global health. A physician explains what it means for doctors and patients.
Vibe-Coded Mashup Mods Have Taken Over X This Week, And They Threaten to Completely Change the Modding Scene
I have spent the last three days scrolling through some of the most insane mod videos I've ever seen. Modders are taking games that have been decompiled with Claude, and mashing them up with other games. As the week has gone on, there have only been more clips of these mods pouring out on X/Twitter.
'This Matters': Researchers Identify Thousands of New Tells in AI Writing
The days of freaking out over em dashes are over.
'This is what happened with Claude and me': AI's failed attempt to crack the Riemann hypothesis led mathematician to a breakthrough
An attempt by Anthropic's Claude failed to prove the Riemann hypothesis, but it did produce something genuinely useful: inspiration.
September Monthly Review: Health MCP Guy?
For all the time I spend talking about APIs and data exchange, this newsletter has had embarrassingly little interoperability of its own. It’s old-school SaaS, baby - UI or bust. Not all of that is my fault! Substack, for all its popularity, is (as mentioned before) really frustrating as my system of record. The primary user it optimizes towards is the reader, whose attention feeds the app, the Notes feed, and the recommendation network that Substack is actually selling. Health API Guy is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber. Subscribe They’re falling into the same trap that Medium once sprinted into, albeit at a much more glacial speed. Younger generations may not remember (or never knew) it, but Medium once lured publications like The Ringer and Signal v. Noise onto its platform, only to lose them as it rebuilt itself around a paywall and an algorithmic feed. Substack’s original pitch circa 2020 was the antidote (you own your list), but every new feed feature chips away at it. I get it: writers are generic supply in the eyes of the Substack product team. Generic supply follows demand, so the consistent pressure on the PMs is to generate demand, which leads to every product cycle going toward discovery (Notes, recommendations, the app, live video). The end result lackluster author tools (laggy and broken analytics are truly crippling) all despite the premium price point. The archive is where it crosses from authorial gripes to actual reader impact, though. Across Substack and LinkedIn, I’ve published 1,218 posts and nearly 900,000 words since 2018. Substack’s native search can see only the 464 posts that live on Substack, and even there it’s woefully inadequate at surfacing relevant content (even for the guy who wrote it)! Readers have the same problem, writ large. Ask the Archive Fixing search (for both my readers and my own use) turned out to be extremely non-trivial, unfortunately: There’s no public API for readers to access content RSS exists for new content notification, but it only carries the 20 most recent posts and cuts paid ones off at the preview The only Substack MCP is an admin tool for analytics The export exists for authors, but it’s painfully awkward and slow. So I built something to work around that: the Health API Guy MCP. It’s a private MCP server holding the full text of everything I’ve written across Substack and LinkedIn. Add it to Claude or ChatGPT as a connector and you can ask things like: It pairs keyword search (for acronyms and case names, of which we have many) with semantic search. Every post is also labeled by topic, form, and court case, so you can pull a lawsuit’s (or multiple cases’) entire docket across Substack and LinkedIn in one query: Hail, Healthapiguyhydra This is also a bit of me eating my own cooking. I’ve spent the year arguing that headless healthcare is arriving and that developers want primitives over abstractions. Last week in Hail, Benihydra, I watched Salesforce plug its platform into Claude and bet that the records underneath outlive whatever interface sits on top: Salesforce must believe that the Hydra’s immortal head is the records, permissions, and business logic (in that they will survive whatever happens to the interface). It felt a bit hypocritical to keep my own archive behind a Substack search bar. The archive is the body. Substack is one head, and now whatever tool you might use is another. Will the EHRs follow my lead? Probably not imminently, but one can hope! Why Founding Members Only The MCP is available to Founding Member subscribers. A few reasons: It costs money and time to run. Hosting, the database, Clerk usage and AI compute for search and labeling are significant operational costs (that increase with usage). Substack gives me no way to automate access. There’s no API to check whether someone is a paid subscriber, so every user is manual operations for now. It’s a beta. We’ll need to improve things and parts will break. I’d rather they break for a small group of dedicated readers who will tell me about it. The Founding tier deserved a real benefit. Until now it was mostly a generous tip. Now it comes with something. If you’re already a Founding Member, reply to this email with the address you’d like provisioned. If you’re not and want in, upgrade here. Larger teams (such as companies) can get access too: group subscriptions at the Founding tier come with the group discount. The admin for the subscription can reply with the list of addresses to provision. Get 20% off a group subscription Pricing Founding membership is now a flat $200 a year, replacing the old suggested $150 that readers could adjust. Existing Founding Members will have access grandfathered in at their previous rate. Month in Review Articles Published Make Networks Dumb Again (Sep 11): AI is making old analog networks cheaper to use and flooding them with machine traffic in the process. A plea for agent-native rails with the intelligence at the endpoints, and for Washington to drag them to ubiquity. Interoperability, Unbuttoned (Sep 29): Interoperability is a series of workflows in a trenchcoat, and this post finally takes the trenchcoat off. A color-coded map shows which provider workflows went digital, which are fragmented, and which still run on fax. Video Content Antitrust’s System of Record Problem (Sep 3): The FTC’s investigation into Epic prompted a ninety-minute webinar, now posted with chapters and a smoothed transcript. We get into why market definition, missing B2B share data, and the SSNIP test leave antitrust poorly suited to systems of record, plus why information blocking may be doing more than any pending lawsuit. The Information Exchange: Cerner Milking Edition (Sep 18): This was so fucking fun. Ryan Tucker, Ryan Brickner, Brad Thorson, and I try an Around the Horn format, complete with points, penalties, and a mute button. We debate whether Oracle Health is dead, who would actually buy Cerner (IBM enters the chat), p(Doom), and why the front desk still hands you a clipboard. Regulatory Ask Not What Your Network Can Do (Sep 23): CDC’s RFI on public health data intermediaries lists use cases already served, unevenly, by eCR, lab reporting, IZ Gateway, and syndromic surveillance. The real opportunity sits in one question about AI: a network that can handle the next emergency without another interface project. Court cases Particle v. Epic: The Naughty Schoolteacher (Sep 4): After a summer-long vigil, the answer on Epic’s early summary judgment bid is two paragraphs and a remedial assignment. Particle advances to the next grade, and market definition gets another year of discovery. The Epic Discovery Files (Sep 8): Texas’s motions to compel arrive with a 251-page appendix of requests, objections, and meet-and-confer letters with Cravath. Inside: Epic using generative AI for document review, denied interoperability requests in scope, and Noerr-Pennington raised against lobbying discovery. Epic v. Health Gorilla: “Working as Designed” (Sep 9): Epic wants nothing to do with the data breach MDL, and explaining why puts fresh discovery on the public docket. The spiciest allegation: Health Gorilla told GuardDog to scrub its law firm business from its website instead of cutting off access. Veeva v. Epic: Legal Bypass Surgery (Sep 15): Veeva asks the Wisconsin Supreme Court to skip the Court of Appeals, while law professors and three policy groups line up as amici. Meanwhile, Epic’s lawyers apologize for another round of citation errors, including a quote that never existed. Texas v. Epic: Behind the Black Bars (Sep 17): The court portal served up Epic’s unredacted filings, revealing a $5 million document review budget and the AI workflow behind it. Also inside: the 200-plus company names Texas wants searched, and one very expensive Zoom invitation. OpenEvidence v. Doximity: Consequences Will Follow (Sep 25): OpenEvidence tells the court it will not comply with a discovery order, and its own lawyers at Quinn Emanuel say they won’t defend the choice. Add years of Slack reported unrecoverable, and Rule 37 sanctions come into view. Health Tech Keeps Choosing Violence (Sep 28): Doximity moves to end OpenEvidence’s case, Epic asks Texas to show its work, and a sleepy Audacious Inquiry patent fight sprouts antitrust and information blocking counterclaims. The main event is a long-sealed Cognizant v. Infosys ruling that hints where system of record antitrust claims can survive. EHRs One Verb for Oracle Health (Sep 14): Oracle’s earnings call quantifies every data center to the decimal point, while Oracle Health gets a single verb: “accelerate.” Panning the few healthcare mentions turns up an agentic care management system, a research-to-care ledger play, and the question of whether healthcare gets Larry’s patience. Epic’s Forretress Under Siege (Sep 30): Judy Faulkner tells a panel Epic paused hundreds of projects to harden its software and calls it a “shame,” while a spokesperson insists the roadmap hasn’t changed. Project Glasswing, a Pokémon-branded QAN sprint, and a trove of Jodel grievances reconcile the two Epics and preview the Jevons paradox coming for all software. Industry Analysis Scanning for a Network (Sep 1): Scan.com raises $220M to build the imaging network labs never got, speedrunning Zocdoc’s consumer-to-B2B pivot. Its wedge to network density may be the most American one available: personal injury litigation. The Front Desk Wants the Damn Whole Building (Sep 16): Hello Patient buys Converse Health, pushing its AI front desk into referrals, chart work, and prior auth. The copilot categories keep blurring, and your distribution partner is still your final boss. Cross-industry Comparisons A Different Patchwork Quilt (Sep 10): Healthcare loves to envy open banking, but the US version has no mandate, a toll booth at Chase, and data quality problems of its own. The grass isn’t greener; it’s another American patchwork quilt, and perhaps a worse one. Hail, Benihydra (Sep 21): Salesforce brings its platform into Claude, Slack, and its own agent, inviting the interfaces that might replace it. Cut off one head and more grow back, so long as the records underneath stay put. Living on A Trade Secret Prayer (Sep 22): Six credit unions are suing Fiserv, the Epic of core banking, by claiming their own member records as trade secrets. Banking has no Cures Act, so its lawyers reach for legal alchemy that healthcare has already tried. Other News None this month External Media Nabla Accelerate NYC: I kicked off fall conference season at Nabla’s Accelerate event in New York, where Delphine Groll opened by asking the room to picture the company it wants to become and the one it doesn’t. I liked her framework of bureaucracy, fear, and ego as the blockers. I also love vendor conferences/events in general - different vibe than industry biggies. Healthcare 101 with Nikhil Krishnan: I briefly guest lectured Out-Of-Pocket’s Healthcare 101 class. Appreciated Nikhil’s takeaways cover whether AI doctors can query HIEs, why healthcare’s data-liberation rules beat most industries’, and the hi-res graphic is in the comments. HCN: Epic Probed by the FTC: I joined the HCN guys to talk through the FTC’s investigation into Epic. Cedric’s new hairdo was an unexpected plot twist. The episode also covers MFN drug pricing, healthcare M&A and Eli Lilly’s acquisition streak. Fall Conferences: At eHealth Exchange’s Annual Meeting in Austin on Oct 27, I’ll moderate a panel on the tensions between AI, HIEs, and timely patient access with Jean Ross (Primary Record), Therasa Bell (Kno2), Byron Crowe (Doctronic), and Michael Marchant (Freenome). Here’s the rest of my schedule if you want to meet up: CommonWell (Redwood Shores, CA): Oct 13-14 Will be participating in a fun Jeopardy event Open@Epic (Madison, WI): Oct 20-23 Will be enjoying Verona, Willy Street, and other local adventures eHealth Exchange and Sequoia Project (Austin, TX): Oct 27-29 Yeehaw HLTH (Vegas): Nov 15-18 Groundhog Day, Venetian Edition RSNA (Chicago): Nov 30-Dec 2 See me at “Governing AI in Breast Imaging: What Health Systems Are Actually Doing” Digital Health Counsel 2026 AI Summit (Seattle): December 2-3 Will likely be speaking to a bunch of lawyers about information blocking Posts I Liked ChatGPT’s Epic integration: game changer or nothing burger?: Joshua Liu, MD offers five thoughts on OpenAI’s Epic launch, starting from the point that it rides the same open APIs available to anyone. He argues read-only access leaves ordering to Art and the ambient startups, and asks why HCA is piloting it alongside Commure. He’s sorta a must-follow at this point - consistently excellent, thought-provoking takes. Epic, Lilly, and the Discovery pass: Seth Chaney reads Epic’s first major Discovery deal (with Eli Lilly), as well as its revised developer terms on commercial displays, as the pipe finally opening between EHRs and pharma. Life sciences and health continue to collide. Scott Rossignol on EHR payer platforms and ePA fees: Scott argues the top ambulatory EHR vendors are turning the prior auth mandate into a per-member-per-year toll on payers, for an API whose operating cost has nothing to do with covered lives. His fix is two lines of regulatory text: put the CRD/DTR/PAS certification criteria in the Base EHR definition, and bring payer-facing ePA connectivity under § 170.404’s cost-based fee rules. It’s a good solution. Health API Guy is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber. Subscribe
Claudette Barreto Lopez, M.D., Joins Baptist Health Primary Care as a Family Medicine Physician
Septmeber 8, 2026 – Family medicine physician Claudette Barreto Lopez, M.D., has joined Baptist Health Primary Care. Dr. Barreto Lopez provides comprehensive primary care with an emphasis on preventive medicine and health maintenance. Her clinical interests include women’s health, nutrition and lifestyle medicine, and she works with patients to incorporate sustainable lifestyle changes that support […]
Gambit's Best of New Orleans 2026: Entertainment
Design by Dora Sison / Gambit Best Bounce Artist googletag.cmd.push(function() { googletag.display('ad-1492300'); }); 1. Big Freedia (Instagram: @bigfreedia)2. Ha Sizzle (Instagram: @hasizzlethevoice)3. Subtweet Shawn (Instagram: @subtweetshawn1) Trixie Minx Photo by Sophia Germer / The Times-Picayune Best Burlesque Performer1. Trixie Minx (Instagram: @trixienola)2. Bella Blue (Instagram: @bellablueforever)3. Miss Angie Z (Instagram: @missangiez) Best Casino1. Caesars New Orleans (8 Canal St., 800-427-7247; caesars.com/caesars-new-orleans)2. Treasure Chest Casino (4550 Williams Blvd., Kenner, 504-443-8000; treasurechest.com)3. Boomtown Casino (4132 Peters Road., Harvey, 504-366-7711; boomtownneworleans.com) Best Drag Performer1. Debbie with a D (Instagram: @debbiewithad)2. Katrina Waters (Instagram: @katrinawaters504)3. Laveau Contraire (Instagram: @laveaucontraire) Best Festival (that's still around)1. New Orleans Jazz & Heritage Festival (nojazzfest.com)2. French Quarter Festival (frenchquarterfest.org)3. Hogs For The Cause (hogsfest.org) Best Food Festival1. New Orleans Fried Chicken Festival (friedchickenfestival.com)2. Oak Street Po-Boy Festival (poboyfest.com)3. New Orleans Jazz & Heritage Festival (nojazzfest.com) Tipitina's Photo by Sophia Germer / The Times-Picayune Best Live Music Venue1. Tipitina’s (501 Napoleon Ave., 504-895-8477; tipitinas.com)2. The Broadside (600 N Broad St., 504-766-0871; broadsidenola.com)3. The Fillmore (6 Canal St., 504-881-1555; thefillmorenola.com) Best Live Music Club1. Maple Leaf Bar (8316 Oak St., 504-866-9359; mapleleafbar.com)2. Snug Harbor Jazz Bistro (626 Frenchmen St., 504-949-0696; snugjazz.com)3. Chickie Wah Wah (2828 Canal St., 504-541-2050; chickiewahwah.com)3. Blue Nile (532 Frenchmen St., 504-766-6193; bluenilelive.com) Best Local Blues Artist/Band1. Little Freddie King (littlefreddieking.com)2. Samantha Fish (samanthafish.com)3. Tab Benoit (tabbenoit.com) Best Local Brass Band1. Rebirth Brass Band (rebirthbrassband.com)2. Hot 8 Brass Band (hot8brassband.com)3. The Soul Rebels (thesoulrebels.com) Best Local DJ1. Paul Lakeway (Instagram: @paul_lakeway)2. DJ Soul Sister (Instagram: @djsoulsister)3. ANTWIGADEE! (Instagram: @antwigadee) Best Local Funk Band1. Dumpstaphunk (dumpstaphunk.com)2. Galactic (galacticfunk.com)3. Big Sam's Funky Nation (bigsamsfunkynation.com) Kaye the Beast Provided Photo by Catherine Debuys Best Local Hip-Hop Group/Artist1. Kaye the Beast (Instagram: @kayethebeast)2. La Reezy (Instagram: @lareezymusic)3. Pell (Instagram: @pellyeah) Best Local Movie Theater1. The Broad Theater (636 N. Broad St., 504-218-1008; thebroadtheater.com)2. The Prytania (5339 Prytania St., 504-891-2787; theprytania.com) googletag.cmd.push(function() { googletag.display('ad-1492297'); }); 3. The Prytania Theatre at Canal Place (333 Canal St., 3rd floor, 504-290-2658; theprytania.com) Best Local Music Artist/Band1. Robin Barnes & The Fiya Birds (robinbarnesmusic.com)2. Trombone Shorty (tromboneshorty.com)3. Sweet Crude (sweetcrudeband.com) Best Local Rock Band1. The Revivalists (therevivalists.com)2. Better Than Ezra (betterthanezra.com)3. Cowboy Mouth (cowboymouth.com) Best Wedding Band/ DJ1. The Mixed Nuts (themixednuts.com)2. Bucktown All-Stars (bucktownallstars.com)3. Warm Advice (warmadvice.com) Doreen Ketchens Provided photo Best Local Street Performer1. Doreen Ketchens (doreensjazz.org)2. Jourdan Blue (Instagram: @jourblue)3. Sariyah Idan (ariyahidan.com) Best Local Theater Company1. The Tennessee Williams Theatre Company of New Orleans (504-264-2580; twtheatrenola.com)2. Le Petit Theatre (616 St. Peter St., 504-522-2081; lepetittheatre.com)3. The NOLA Project (504-302-9117; nolaproject.com) Best Local Traditional Jazz Group/Artist1. Preservation Hall Jazz Band (preservationhalljazzband.com)2. Kermit Ruffins (@ruffinsbbq)3. Trombone Shorty (tromboneshorty.com) Best Marching Band1. St. Augustine High School Marching 100 (Instagram: @staugmarching100)2. Edna Karr High School Marching Band (Instagram: @theofficialkarrband)3. The Roots of Music Marching Crusaders (therootsofmusic.org) Best Mardi Gras Parade1. Muses (kreweofmuses.org)2. Iris (kreweofiris.org)3. Endymion (endymion.org) 610 Stompers Photo by Chris Granger / The Times-Picayune Best Mardi Gras Dance Krewe1. 610 Stompers (610stompers.com)2. Amelia EarHawts (aecabinkrewe.com)3. The Jamettes (Instagram: @thejametteskrewe) Best Place to See Burlesque/Drag1. The AllWays Lounge & Cabaret (2240 St. Claude Ave., 504-218-5778; theallwayslounge.net)2. The Original Nite Cap (1300 St. Bernard Ave., 2nd floor, 504-224-2854; Instagram: @originalnitecap)3. The Country Club (634 Louisa St., 504-945-0742; thecountryclubneworleans.com) Best Place to See Comedy1. Sports Drink (1042 Toledano St.; sportsdrink.org)2. The Howlin' Wolf (907 S. Peters St., 504-529-5844; thehowlinwolf.com)3. Big Couch New Orleans (1045 Desire St.; bigcouchnola.com) Best Local Comedian1. Ryan Rogers (ryanrogerscomedy.com)2. Mark Normand (marknormandcomedy.com)3. Cyrus Cooper (Instagram: @cyrusjcooper) Best Strip Club1. The Penthouse Club (727 Iberville St., 504-524-4354; penthouseclubneworleans.com)2. Rick’s Cabaret (315 Bourbon St., 504-524-4222; rickscabaretneworleans.com)3. Larry Flynt's Hustler Club (225 Bourbon St., 504-524-0010; neworleanshustlerclub.com) Best Performing Arts Center1. Jefferson Performing Arts Center (6400 Airline Dr., 504-731-4700; jeffersonpac.com)2. Saenger Theatre (1111 Canal St., 504-525-1052; saengertheatreneworleans.com)3. Mahalia Jackson Theater for the Performing Arts (1419 Basin St., 504-525-1052; mahaliajacksontheater.com) .donate-gambit-article-module-container { text-align: center; padding: 10px; /* background-color: #24da6d; */ background-color: rgb(153, 0, 0); margin-bottom: 20px; } .donate-gambit-article-module-title { font-weight: bold; color: #fff; } .donate-gambit-article-module-copy { color: #fff !important; font-size: 16px; } .donate-gambit-article-module-button { background-color: black; width: 50%; border-color: black; font-size: 16px; } .donate-gambit-article-module-button:hover { background-color: black; } @media only screen and (max-width: 600px) { .donate-gambit-article-module-button { font-size: 14px; width: 100%; } .donate-gambit-article-module-copy { font-size: 14px; } } BECOME A GAMBIT MEMBER AND JOIN OUR KREWE TODAY For more than 40 years, Gambit has covered New Orleans, for New Orleans. We’ve stood up to city hall with you, cheered the Saints with you, danced in the streets with you and cried over our collective loss with you. And we’re plannin’ on being here for 40 more. To do that, we need your help. Become a Gambit member today and help us continue our mission SUPPORT GAMBIT
Trump unveils new site to simplify access to government services
America.gov for now functions like a chatbot comparable to ChatGPT or Claude, pointing users to official gov't websites.
Did Anthropic’s A.I. Really Make a Scientific Discovery on Its Own?
An expert at the University of Copenhagen said his team had been sharing its research with the company’s A.I. model, Claude, and that its new finding matched their work.
Why letting Claude clean your TV's bloatware isn't the best idea
Don't let an AI bot potentially crash your TV.
Anthropic says Claude found a new gene-editing enzyme but science isn't confirmed
Anthropic says Claude agents autonomously flagged a CRISPR-like enzyme system in virus DNA in a 21-hour run. A CRISPR pioneer calls it intriguing, others sceptical — it's an unreviewed preprint with function unknown.
US court upholds Pentagon’s blacklisting of Anthropic
Conflict began in February when Anthropic refused to remove restrictions on use of Claude for fully autonomous weapons.
Claude spots a mystery enzyme scientists overlooked
Autonomous agents flagged a family of phage enzymes with CRISPR-like repeats, though what the system does remains unknown.
BizTrip AI Announces Connection With Claude and ChatGPT
Users will be able to manage travel without switching to separate app or website
BizTrip AI Brings Business Travel Booking Directly Into Claude and ChatGPT
SAN FRANCISCO, September 23, 2026--BizTrip AI, makers of agentic AI for corporate travel management, today announced that its core trip-planning and booking experience is now available natively inside both Claude and ChatGPT. The integration uses the Model Context Protocol (MCP) — specifically MCP "apps," which support rich, interactive rendering directly in a chat conversation — allowing travelers to search, compare, and book business travel without switching to a separate app or website.
The legendary Hi-Ho Lounge will return to St. Claude Avenue following closure of No Dice
A soft opening is already planned this month. Here’s everything you need to know.
Dealroom.co | Anthropic pours A$10M more into Australian science, lifting total to A$13M
What's the deal? Anthropic is adding A$10 million in Claude API…
This Ford exec's family Claude agent is on a PIP
Whitney Stefko Dover built a family chief of staff with Anthropic's Claude. The agent, named Claudette, has fresh competition for her job.
Business Tech News: 86% Of Solopreneurs Try AI Before Hiring
SurveyMonkey, Autodesk, Claude, QuickBooks and FreshBooks roll out new AI, security and support tools aimed at helping small businesses work smarter.
Anthropic Reportedly Delays IPO; OpenAI Expects Massive Cash Burn
AI startup Anthropic is now targeting a November IPO, the Wall Street Journal reported.
Claude Code relaunches Projects to manage multiple AI agents in the cloud
Claude Code can run teams of AI agents in one spot now.
Anthropic launches Claude Code Projects, an ‘always-on’ conversation that remembers and delegates your long-running dev work
For enterprises, it makes a whole lot of sense: their digital storefront, website, content management system, procurement platform, or other business application rarely has a discrete endpoint.
Google Home MCP lets Antigravity, Claude, OpenClaw, & more control your smart home
Google today announced “Home MCP” to let your personal AI agent interact with your smart home devices via the Model Context Protocol.
Your AI agents can now control your Google Home devices
Google is launching early access to a new MCP server for Google Home, allowing AI agents like Claude, ChatGPT and others to control connected devices, review camera summaries and access smart home activity using natural language.
Anthropic debuts Claude Docs, raising stakes for Microsoft
Frontier AI labs are unleashing products that compete with established software giants and startups alike
Anthropic merges its Claude chat and agentic Cowork products into a single AI assistant as part of a push to build an AI superapp
Anthropic is folding its chat and autonomous task-execution products into one Claude interface, while adding collaborative document and presentation tools.
Claude comes for Gemini with its own take on Docs and Slides
Claude’s getting its own version of Google’s productivity tools.
Anthropic merges Claude chat and Cowork in one interface
Anthropic is releasing these features to Pro and Max plan subscribers initally
Novo signs Anthropic deal to test Claude Science in R&D
Anthropic partners with Novo to tackle drug discovery using Claude Science AI. Financial terms undisclosed. Novo also has a similar deal with OpenAI.
Novo Nordisk Taps Anthropic’s Claude Science To Speed Up Drug Discovery — Here’s Everything Investors Need To Know
Novo will use Anthropic’s Claude Science in R&D to tackle scientific and drug discovery challenges, aiming to develop new medicines faster.
Novo Nordisk Taps Anthropic’s Claude Science To Speed Up Drug Discovery — Here’s Everything Investors Need To Know
Novo will use Anthropic’s Claude Science in R&D to tackle scientific and drug discovery challenges, aiming to develop new medicines faster.
Expensify in Integration With Anthropic
Expense management provider one of first to connect with Claude for Small Business
Meta now lets AI agents handle the boring parts of WhatsApp Business setup
A new WhatsApp Business MCP server lets developers use AI coding agents like Claude, Cursor, Codex, and ChatGPT to handle setup, messaging templates, testing, and troubleshooting.
Google finally lets all engineers use Anthropic's Claude
Google engineers can finally use one of Anthropic's strongest coding models for internal work.
Apple's Siri AI Can Be Swapped Out for Claude, ChatGPT, Code Shows
Code sleuth "pdfu" has uncovered iOS 27 and macOS Golden Gate private frameworks that show Apple has designed its new Siri architecture to work with third-party AI models at what appears to be a surprisingly deep level. One mechanism called Model Delegation allows Claude to appear as a Siri extension in the same way as the existing built-in ChatGPT extension.
China AI labs routed requests to Claude 35 million times: Anthropic
Anthropic said Alibaba, Moonshot, Zhipu, DeepSeek, and Xiaomi all tried to use Claude to train their models in various ways.
Anthropic says it disrupted scientists using Claude AI for possible biological weapons development
Anthropic has cute graphic showing how its AI spread 'malicious' code
Anthropic said it was "most concerned" about an event in which Claude uploaded "malicious" code. To help explain the incident, here's a cute robot.
Viewpoint: Unlocking Europe’s wildfire data
Europe is already investing in the science, technology and infrastructure needed to understand, prevent and fight wildfires. Copernicus provides satellite-based Earth observation and environmental information, while the European Forest Fire Information System provides information on fire danger and forest fires across Europe. Weather and climate systems add forecasts of heat, drought, wind and humidity; historical fire databases show where fires have occurred; and researchers generate increasingly detailed knowledge about fuels, fire behaviour, forecasting, modelling and risk management. Meanwhile, new European research projects are adding digital twins and other tools to this growing ecosystem.However, these data do not necessarily sit together, nor are they necessarily described, structured or governed in ways that allow them to be used together. Much of the information is public, but some is commercially held. Some is sensitive. Some is subject to legal, security or operational restrictions. Even when data can be legally accessed, different systems may use different formats, terminology and rules.Europe therefore faces an unusual paradox: we have vast amounts of relevant information to respond to the wildfire challenge, but not the means to bring it together. For example, a wildfire researcher or emergency service might need to ask whether a particular combination of vegetation, drought, wind and topography has historically been associated with rapid fire spread. The relevant information may exist in several institutions, using different systems and subject to different conditions of access.The challenge is to conduct an authorised analysis that interacts with these sources, combines the results and respects the rules imposed by each data custodian.An important part of the answer can be found in the Fair principles, the idea that data should be findable, accessible, interoperable and reusable. Developed more than a decade ago, these principles have become a central reference point for modern data stewardship, making it easier for people and machines to find data, understand what it means and determine whether it can be reused.Readability by machines is essential, since AI systems will be key to meeting the data challenge. Bring analysis to the dataThe proposition is not to give an AI system unrestricted access to sensitive databases. Quite the opposite. The goal is to make it possible for approved algorithms to go to the data, rather than requiring sensitive data or large datasets to be copied to wherever the analysis happens.Imagine an analysis asking a series of questions of different data stations. At one station it might examine weather and climate data; at another, historical fire records; at another, experimental measurements of fire behaviour. The algorithm does not need to download everything. It can operate within the environment permitted by each data custodian and return only the results that the rules allow to leave.This is sometimes described as data visiting, and related approaches already exist in areas such as federated analysis and federated learning.Sensitive data should therefore not be treated as a free-for-all for machines. This requires appropriate governance at multiple levels. Access needs to be authenticated and authorised. The purpose of an analysis should be defined. The code or model being run should be controlled. Data custodians should be able to specify what can be queried and what results can leave their environment. Activities should be auditable and human oversight should remain where the consequences require it.In other words, the objective is not to trust AI with the data. It is to design systems in which AI never has more access than it needs.As open as possible, as closed as necessaryFair data should also not be confused with open data. Some wildfire data should be openly available. Other information may legitimately need to remain restricted because it concerns critical infrastructure, security, commercial interests, privacy or operational capabilities. The principle should be: as open as possible, as closed as necessary.The important point is that restricted data should not automatically become unusable data. If machines can discover what information exists, understand its meaning, establish where it came from and determine the conditions under which it can be used, authorised analyses can potentially be carried out without the underlying data ever leaving the environment in which it is protected.Restricted data can still be Fair and machine-actionable.This is also an important question of data sovereignty. European institutions should not have to surrender control of their data to a handful of technology platforms in order to benefit from AI. Nor does Europe need to start again by building one giant repository for every piece of wildfire information. The better approach is to connect what already exists.This is the principle behind emerging approaches such as Tryangle, developed through the Leiden Initiative for Fair and Equitable Science. Rather than creating another centralised data warehouse, the approach seeks to provide an interoperable layer through which approved algorithms can visit distributed data sources while respecting the rules of their custodians.In this way, a data station can remain where it is. Its custodians can retain control. Existing databases and applications do not have to be abandoned. What changes is their ability to communicate with other systems through common, machine-readable descriptions, access rules and interoperability mechanisms.But approaches like this need data that meets the Fair principles. Only then can an interoperable ecosystem safely begin to turn fragmented information into collective intelligence.Data infrastructure is wildfire infrastructureThe implications of this approach are entirely practical. Firefighters could gain better intelligence about local and evolving risks. Land managers could test prevention strategies using more comprehensive information about fuels, weather and past fires. Researchers could compare evidence across ecosystems without repeatedly moving huge datasets between institutions. Smaller regions and resource-constrained teams could access analytical capabilities without having to build large data infrastructures themselves.Europe is already investing heavily in satellites, databases, research programmes, forecasting systems and scientific expertise. The question is whether we will connect those investments well enough to turn them into actionable intelligence.That requires policymakers and funders to treat data infrastructure as seriously as physical infrastructure. Data generated or supported through European public funding should be Fair, verifiable and machine actionable by design, wherever legally and ethically possible.Related articlesData Corner: Europe’s burning need for wildfire mitigation researchClimate monitoring must remain a global enterprise, experts sayIn practice, that means funding not only the production of data, but also the infrastructure that makes data usable across systems: metadata, provenance, interoperability, access mechanisms and governance. Where data must remain restricted, we should ensure that it can still be securely discovered, verified and used for authorised analysis.Initiatives such as the European Open Science Cloud offer an existing framework through which these principles can be put into practice. What is needed now is to accelerate implementation and make them a standard part of how Europe funds and builds research and innovation systems.We do not need another generation of disconnected databases. The data is already there. The opportunity is to make it work together, securely, responsibly and quickly enough to help prevent the next fire, rather than simply document the last one.Jean-Claude Burgelman is professor emeritus at the Free University of Brussels (VUB) and editor-in-chief at Frontiers Policy Lab. For more than 20 years he held leading roles in research and innovation policy at the European Commission.
An AI Formalized and Verified Fermat’s Last Theorem in 11 Days, a Task Expected to Take Years
Claude did not rediscover the proof. It made the existing one machine-checkable.
Anthropic drops plans to acquire Israeli AI startup Decart
The companies had discussed a transaction worth roughly $6 billion, but the Claude maker ultimately decided against buying the company after reviewing its business and technology.
Anthropic's Labs team drives AI innovation as IPO nears
Anthropic Labs built some of the company's biggest hits, like Claude Code. Cofounder Ben Mann lays out his team's efforts and changing ambitions.
Anthropic’s Claude Can Now Autonomously Run Science Experiments With Lab Equipment
A new system allows agents to orchestrate complex experimental processes and extends Anthropic's reach into the physical world.
Anthropic upgrades Claude's computer use to run in the background on Mac
While computer use isn’t new for Claude on the desktop, running it in the background is a nice upgrade. Anthropic...
August Monthly Review: ChatGPT In Epic
Death, taxes, another frontier lab healthcare launch inspiring truly insane LinkedIn takes. As much as I want to stay away, I repeatedly am the crewman unplugging his ears whenever we pass these sirens. So it’s not the first and probably not the last time we dial up the typewriter rather than tie ourselves to the mast. The full blow-by-blow timeline: Health API Guy is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber. Subscribe The End of the Standalone PHR (Jan 8): ChatGPT Health arrives for consumers, at the time powered by b.well. I was bullish on the approach in that the best way for the PHR to finally solve distribution was to meet consumers were they already are (i.e. When Horizontal Meets Healthcare (Jan 9): OpenAI for Healthcare puts out the shingle for enterprises, with SharePoint ingestion standing in for the EHR integration the product actually needed. Another One: Anthropic’s Healthcare Debut (Jan 16): Claude's version, on HealthEx rails instead of b.well, with Agent Skills for FHIR development as the one true vertical-specific investment. ChatGPT for Clinicians: The Trap Sprung (Apr 24): OpenAI filled in the missing GTM middle with a free product for verified clinicians, effectively peeling the enterprise wrapper off ChatGPT for Healthcare and taking the PLG fight directly to OpenEvidence. OpenAI’s Second Attempt at Health (Jul 31): a relaunch of their patient specific app, now branded Health in ChatGPT, where the privacy partition came out, the connectors thinned, and b.well disappeared in favor of first-party FHIR work against a smaller network Beyond a Shingle So what was announced now? Today, we’re introducing a new electronic health record integration that brings authorized patient context from Epic into ChatGPT for Healthcare, along with the Healthcare Public Data plugin for direct, structured access to official healthcare datasets like PubMed, DailyMed, and CMS Coverage. Together, these capabilities bring ChatGPT closer to the systems and sources healthcare teams trust, while supporting the controls and compliance healthcare work requires. The core portfolio is unchanged from prior announcements we delineated and discussed (aside from the relaunch of patient-facing). Since trade publications still seem to be confusing them by using the wrong names: Enterprise sales motion: ChatGPT for Healthcare Protecting their core product / PLG: ChatGPT for Clinicians Consumer/patient-facing product: Health in ChatGPT So really, this is a feature release (specifically for ChatGPT for Healthcare when it comes to EHR integration). What’s cool about it, though, is that we are moving beyond just putting out a shingle. Everything in the January enterprise launch was horizontalizable: Horizontal tech companies thus generally start verticalization with their “putting out a shingle phase”. Telling the world you’re open for business as a horizontal tech company reliably attracts early inbound interest from buyers who are already trying to force-fit horizontal tools into domain-specific workflows and are eager for any signal that the vendor intends to support their use case more directly. Nothing was really vertical specific, which is exactly what you'd expect from the ultimate hypergrowth horizontal company dipping their toes into specific industries. But eventually you have to stop changing the sign on the door and start changing the product. Can you guess where the horizontal product was going to run out of road? This is the core dilemma writ large: verticalization requires differentiation, while scale economics of consumer and horizontal push toward unification. I’ve buried the lede so deep here it will be painful to some readers, but what should be shocking to no one is that my perspective is that EHR integration is the only path that meaningfully resolves this tension for ChatGPT for Healthcare. Integration with Microsoft Sharepoint is fine, but it is categorically insufficient for any product that hopes to influence clinical decision-making. SMART Money Well here we are, eight months later! Real vertical investment has begun, which surprised a few people. One friend in the industry messaged me: “Absolutely no way Epic gave ChatGPT API access.” Looking forensically at the announcements, the integration is notably read-only across appointment notes, laboratory results, medications, and specialist documentation. So while I’m not a betting man, I'd wager the house this is a SMART on FHIR launch: That sounds a lot like USCDI! The recommended path for the workflow they outline would be SMART on FHIR Their Health AI lead’s post is evocative of both standalone and EHR launch SMART, which another termed “ChartGPT” and “EHR Plugin” The implementation burden for SMART on FHIR is the lowest of the available integrative paths This is the team who literally did patient-facing SMART on FHIR with Epic last month Most importantly, if you’re a horizontal company, you generally still haven’t taken the full plunge and thus (overly) value reusability. SMART on FHIR is the logical compromise: healthcare-specific enough to matter, but standardized enough to reuse. Nobody Asked Judy Another friend asked “I thought they had to go to each health system, or did they go direct to Epic and now health systems opt in?” Given the trust paradigms in healthcare, this functionally cannot be Epic hoovering up all the data for OpenAI and shoving ChatGPT in their customers’ faces (or they would revolt). However, I do believe the aforementioned terrible coverage by TechCrunch and other outlets is probably responsible for that question, given the 325 million reference. When using FHIR as a business associate to an Epic customer, there is no EHR gatekeeping to speak of, as that isn’t how fhir.epic.com works: You register an app You pick your APIs You test against the sandbox APIs You list as “Ready for Production” Hospitals pick your app You test and go live with them Bluntly, nobody had to say yes or no in Verona (nor were they given the chance). Vendor Services (their next developer tier up that I doubt OpenAI is using quite yet in this initial release) certainly has contractual paperwork that some resent, but even there, the twin pressures of mounting antitrust and information blocking make outright gatekeeping increasingly fraught. My friend’s sentiment is, to me, representative of a broader industry neurosis, perhaps a sort of scar tissue of prior eras: people assume the gate is still there and never actually try the door. There are certainly other ways Epic (and any EHR) can put its thumb on the scale, but they are not so dumb as to stand in front of a federally mandated API and play bouncer at this exact moment in time. So I think they can and probably should go deeper, as this release is at best parity and at worst behind vertical-specific competition. OpenEvidence did a basic SMART launch with Sutter Health in February, allowing for more convenient evidence search by providers. They then add patient-context (the equivalent SMART on FHIR flow to what we see here) when Cedars-Sinai went enterprise-wide in May. UpToDate has four distinct applications across deeper workflows like patient engagement listed in Epic Showroom and announced a partnership with Epic to power Art at UGM Abridge and other ambient scribes have invested deeply into the deepest bidirectional clinical copilot workflow with integrations well beyond SMART on FHIR. In that light, the “Who is OpenAI primarily targeting here?” is clear. This release brings them into striking distance of OpenEvidence, but not the others quite yet. They are the only one of the three you can reach without building something truly Epic-specific. When will they go further? Competition is a great motivator to overcome the horizontal demons and start building things that can’t be reused: not across industries, not across EHRs, maybe not past the customer you built them for. That’s the price of actually verticalizing. The question is if and when OpenAI will be willing to pay it. Month in Review Here is the monthly review. As a reminder, this is a regular round-up of the month’s posts and other content to surface things you may have missed across regulation, litigation, interoperability, and beyond. AI assistance is used in these bullet summaries so I can focus on articles. Articles Published: None this month Video Content: The Information Exchange: Standards-based Thruple Edition (Aug 14): Back to school for us too, with Brad reporting in from the CMS Health Tech Ecosystem’s one year anniversary in DC. We get into the January CMS-0057 deadline, the fall rulemaking reading list, and why enrollment still keeps most apps off FHIR. The Information Exchange: Epic Dúnadan Edition (Aug 25): A UGM roundup with Ryan Brickner joining for the first time to check our takes against what the building actually thinks. Rangers, Ergo sitting on top of EHI, and a down-market lineup that needs some cuts. Regulatory: Stacked Deck, Bad Hand (Aug 05): ONC’s website refresh took the entire HIT Policy Committee record with it, so I rebuilt the archive and went looking for the regulatory capture story everyone assumes is buried in there. Stacking the deck and winning the hand turn out to be very different things. The Sixth Generation of Patient Access (Aug 19): A quick catalog of the five generations of patient access we’ve layered on since HIPAA, and the two candidates now competing to be the sixth. One is planned. The other is the market routing around the plan entirely. Court cases: Three Cases Walk Into a Docket (Aug 06): Two surprise settlements cleared the board in a single week, and then Judge Maddox dropped ninety-four pages on Vyne v. Henry Schein. The sleeper is a DMCA holding that makes direct-to-database a considerably riskier business model. Epic v. Health Gorilla: Into the MDL (Aug 10): Nine class actions are headed to Miami, and the Panel signaled it wants to bring the case that spawned them along too. That would leave Epic arguing the requests were obviously fraudulent in one courtroom and unknowable in the other. Amazon v. Perplexity: Agents Are Legalized! (Aug 11): The Ninth Circuit vacated the injunction against Comet, holding that the user is the one accessing the servers rather than the company that built the agent. A real win for agentic access, and a much narrower one than the headline suggests. Veeva v. Epic: Come At Me, Bro (Aug 26): Epic’s response brief wants the dismissal affirmed and, unusually, wants the opinion published as precedent. Buried in it is Epic’s own description of what its non-competes actually prohibit, which current and former employees should read closely. EHRs: The Contract Epic Would Never Sign Today (Aug 04): A 1999 SEC exhibit catches Epic licensing nearly its entire product line, Tapestry included, to the company that became TriZetto. The marketing services menu attached to it is the part that will make you blink. Forecast from Verona (Aug 07): The Epic Almanac makes one argument six different ways: the AI is only as good as the networks behind it. Emmie headlines, Art’s context stack is the most ambitious part, and Penny finally gets an autonomous coding date on the calendar. Works With Epic MyChart (Aug 17): Epic’s first new Showroom category since the death of Workshop certifies hardware instead of software, badge on the box and all. Made for iPhone, but for blood pressure cuffs, and rough news for anyone selling the RPM stack sitting in between. UGM Hot Takes 2026 (Aug 20): Everyone else covered the AI announcements, so I went after Savvy deleting the payment gateway, Rangers as Boost with the timer removed, and a down-market lineup with too many entries. Plus the Health Grid tidbits I cannot help myself on. MyChart Central Grows Up (Aug 24): Device data and Emmie turn Epic’s identity hub into a full consumer platform, which is both a logical answer to ChatGPT and a bit of a mistake. It also happens to be the best scraping target Epic has ever shipped. Industry Analysis: The Wrong Yardstick (Aug 12): Vertical software keeps getting judged against Superhuman and Notion, which is the wrong bar entirely. Your user is comparing your product to a whiteboard and forty phone calls before lunch. Primitives vs. Abstractions (Aug 27): Developers want building blocks, systems of record prefer to hand out business logic, and both sides have a real case. There’s no test from the outside that separates the engineering reason from the competitive one, which is why this keeps ending up in court. Cross-industry Comparisons: The Two Kinds of Platform Power (Aug 18): Attention power and record power are different problems that keep getting handed the same regulatory toolkit. Congress’s latest swing at Big Tech shows both what’s possible and where it falls apart. The Dogs of (Platform) War (Aug 21): A CourtListener alert turned up a property management fight with healthcare’s exact shape and none of the Cures Act. Shell prospects, ghost accounts, a notetaker bot that came back to haunt someone, and a lawyer arguing both sides of the same theory in two states. Fractals All the Way Down (Aug 31): Yardi dominates property management right up until you zoom into one segment, where AppFolio owns it outright and Yardi doesn’t appear at all. Where you draw the line around a B2B software market is about to decide an awful lot of cases. Other News: If Judy Had the Courage (Aug 13): A short eulogy for the era when software was allowed to look insane: WinAMP, bold colors, and everything the grey chatbot era has taken from us. External Media: Portland Monthly Health Tech Meetup: Erin O’Brien channeled my own feelings - maybe it’s just because it’s August and beautiful, but the monthly edition of the PDX Health community meetup was a ripper. Make sure to reach out if interested to join for the next one. STAT’s Coverage of Epic This Month: Brittany Trang of STAT did a fantastic job of breaking the FTC investigation that’s been lurking, as well as UGM related coverage, so I had to give her her wish of a meme. Fall Conferences: I’m pumped to be kicking off my fall conference season in NYC at Nabla Accelerate next month, which Chrissy provided the link to apply for. Here’s the rest of my schedule in case you want to meet up: Commonwell (Redwood Shores, CA): Oct 13-14 Open@Epic (Madison, WI): Oct 21-22 eHealthExchange (Austin, TX): Oct 27 Sequoia Project (Austin, TX): Oct 28-29 HLTH (Vegas): Nov 15-18 RSNA (Chicago): Nov 29-Dec 3 Posts I Liked: HTI-6 Should Unbundle API Certification: Josh Mandel makes the case that HTI-6 should split (g)(10) into separate authorization and data certifications, so a PACS or a genomics platform can certify only the role it actually performs. Imaging is the urgent example, but the structure solves a much bigger problem. On implementing ePrior Auth with Epic: Scott Rossignol’s field notes from a live ePrior Auth build on Epic, including the two app registrations nobody warns you about and the CPT mapping problem waiting at the end. CMS-0057 implementations are going to be such a beast. Healthcare point solutions are starting to look a lot like streaming services: Spencer Dorn runs the cable-to-streaming arc against health IT and lands squarely on the bundling half of the cycle. Bundling can be good! It can also be bad! Life is nuanced. Health API Guy is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber. Subscribe
Anthropic tightens training security after Claude agents went rogue
Anthropic tightened its AI security and paused some programs after Claude models accessed three organizations' systems without permission in April.
Run NVIDIA BioNeMo NIM Microservices for Protein Structure Prediction in Claude Science
Agentic AI is changing how research is done. AI scientists can read papers, propose hypotheses, call models, and determine which experiments to prioritize next. First proving their value in software…
Xero Adds AI, Google Targets Law Firms And Claude Beats ChatGPT: 5 Small Business Tech Stories
Xero adds AI tools, Google expands Gemini for law firms, LinkedIn fights AI slop, Goldman warns of AI risks, and Claude tops user satisfaction.
Sony says Claude trained on pirated lyrics, seeks $150K a song
In a new lawsuit, Sony Music and Warner Chappell say Anthropic used "thousands upon thousands" of songs to train Claude, including "Eye of the Tiger."
4 Claude Skills Every Data Scientist Needs in 2026
Four skills worth adding to your workflow today if you don't want to be left behind
A photographer spent $20K on drives, then used AI to save thousands
After spending $20,000 on drives in 2025, photographer Allen Cooley used Claude to organize old storage and avoid buying more.
Anthropic makes first move into physical AI with new way for scientists, manufacturers to bring equipment to life
The Model Hardware Standard (MHS) works with any AI model, not just Claude, but for now only a handful of companies can access the research preview.
Blood spatter, bats, beer: This metal bar’s ‘Science Time’ happy hour a hit on St. Claude Ave.
What do you get when you combine science, drink deals and goths? That’s Science Time, baby. Here’s an inside look.
Claude Moore awards $2.7 million for health care scholars program
The Claude Moore Charitable Foundation has awarded $2.7 million in 2026 Claude Moore Scholars grants to 18 school divisions, higher education institutions, health care employers and workforce partners across Virginia,
5 Things To Vibe Code With Claude To Build A Better Business Now
You could vibe code 5 things that could transform your business. Here's what to do and exactly how to do it. No new team member required.
Claude Has More Google Workspace Integrations...And Other Business Technology News This Week
Claude expands workplace integrations, ChatGPT can edit Drive files, AI job losses remain limited, and LinkedIn offers an AI marketing playbook.
Anthropic Claude Adds Watermarks. Implications For Business?
Anthropic just watermarked all Claude text for EU rules. What this precedent means for your business, and specific actions that business leaders can take today.
Claude can do more with Gmail and Google Drive, and access to Claude Cowork is expanding.
A pair of announcements from Anthropic today include the word that access to Claude Cowork on mobile and web interfaces has rolled out to all paid accounts, so you can keep an eye on what it’s doing for you in more places. Separately, the Google Workspace connectors for Claude now support sending emails in Gmail and managing files in Google Drive — more details here. [Image: https://platform.theverge.com/wp-content/uploads/sites/2/2026/08/778273589_17980259259118227_5377280649389143563_n.jpg?quality=90&strip=all]
Claude can now send emails in Gmail, even without your approval
Anthropic’s Claude AI can now take a huge step in your email inbox, with the ability to not only interact...
How Claude is accelerating protein design and analytical chemistry
In this post, we share two results that show how Claude can help life scientists increase the pace of their research. In the first, we tested Claude’s ability to design protein binders from scratch, a key step in creating protein-based drugs that has historically taken a specialist weeks or months per target. In the second example, we evaluated whether Claude can accelerate chemical analysis. Claude Opus 5, a generally available model, was given NMR and LC-MS data (the data that allows chemists to assess the identity and purity of the compounds they work with).
Anthropic confirms Claude is down in major outage affecting multiple services
Claude is experiencing a major outage, with users reporting login problems and degraded performance across several Anthropic services.
This platform gives you ChatGPT, Gemini, Claude, and 17+ other AI models for life for $54.97
20+ AI models for a one-time fee
Alarming study finds top chatbots more likely to repeat falsehoods from the left: ‘Truly astonishing’
ChatGPT, Gemini, and Claude were more likely to fall for political falsehoods from the left, while Grok skewed right, a new misinformation study found.
AI needs science’s search history
“Actually, that’s where the gold is,” said Alasdair Russell, my graduate school friend who leads a pre-clinical genome editing group at Cambridge. He was talking about the winding road of science that is omitted from published papers. “When you’re discussing how to do the experiment, and why this way is better than another way, and what does the data really mean? I know what it shows, but what does it mean?” At RAAIS 2026, he described how his group has begun logging the twisting path of discovery as it happens, recording the verbal and written exchanges that normally disappear. Ideas become nodes in a living graph: they branch when a meeting produces two plausible experiments, merge when separate lines of evidence converge, and go dark when someone quietly stops pursuing them. In his implementation, one agent scores novelty, while another tries to learn “how scientists think and how they navigate through a complex world of data,” so that high-potential nodes can trigger deeper investigation. As frontier AI labs deploy agents toward scientific discovery, giving AI authentic scientific taste remains a trillion-dollar dilemma. The bottleneck is that the record we have kept for centuries of science might be insufficient to get us there. The experiments that never make the paper Peter Medawar, 1960 Nobel laureate and “father of transplantation,” asked in 1963 whether the scientific paper was a fraud, and answered yes: the form of the paper misrepresents the thinking that produced it. Sixty years on, the diagnosis is unchanged. A paper presents a clean progression from hypothesis to result to conclusion. Lost along the way are the unconventional theories, the abandoned or unaffordable methods, and the underwhelming and inconclusive data. It reads like a browser history with every dead end deleted: the ten open tabs, the four rephrased queries, the wrong turn down a subforum. All scrubbed, leaving one clean path from question to answer as the canonical path. What has changed since then is that the discarded material is now worth something. No high-impact journal wants these artifacts, but raw trial and error is the most likely source of the training data required to develop scientific intuition, or what researchers call taste. Earlier attempts to capture the discards were motivated by scientific integrity. The Journal of Negative Results in Biomedicine launched in 2002 to publish rigorous studies that disputed established models or exposed ineffective treatments. Its archive preserved negative conclusions after they had become papers, rather than the live alternatives and arguments that produced them. BioMed Central closed it in 2017, saying the mission had been served now that other journals publish null results. A less generous reading is that in fifteen years it published around 200 papers because almost nobody wants to read a negative result, let alone write one up when it will not count toward academic tenure. AI models are different. Even an uninteresting negative result can be useful, provided it is labeled. Earlier this year, Anthropic put a version of this to the test. The company pulled 129 decision points from real Claude Code sessions between January and March 2026, showed models the work up to a human detour, and asked what should happen next. Claude Mythos Preview beat the human choice 64% of the time, while Opus 4.5 managed 51%. The comparison was tilted toward the models because Anthropic deliberately chose moments where the human decision had room for improvement. On 127 further scenarios where the human action was already strong, the models improved on it only about 20% of the time. The study was possible because Anthropic’s researchers work inside a tool that logs by default. The reasoning, detours and outcome are produced in the same working environment. Biology has no equivalent. Its reasoning happens in hallways, at benches, in Slack threads and on whiteboards, while the outcome arrives weeks later somewhere else. The record has to emerge from the work itself. Ask scientists to reconstruct it afterwards and we’ll create another polished account. Subscribe A log is not a label Before any of this becomes training data, a negative result has to say what failed. The first paper published in the Journal of Negative Results in Biomedicine examined 234 negative studies across five leading medical journals. Only 30% discussed statistical power, and half clearly defined a primary outcome. Tell a model an experiment failed, without telling it whether the assay was underpowered, a reagent had degraded, or the hypothesis was simply wrong, and it will learn noise with confidence. Decision histories have a second missing label too. When a lab considers five experiments and runs one, only the chosen branch returns an outcome. The other four are experimental counterfactuals. Researchers call this the selective labels problem: the data reveal results only for the actions someone chose to take. Such a record can teach a model to imitate a lab’s taste, but it cannot establish that the taste was good. It also smuggles local constraints into general lessons. A lab that never ran cryo-EM because it did not own a microscope teaches a model that cryo-EM is rarely the right call. Alasdair’s nodes preserve the candidate set, which is the necessary first step. To be useful, each decision record needs six fields: The evidence available at that moment, and nothing that arrived later. The candidates considered, including the ones dismissed in a sentence. For each candidate: the expected result, the confidence attached to it, and the cost in time and money. The route chosen, and the reason. What happened next. How the evidence changed the scientist’s view. These must be timestamped before the outcome, because hindsight turns uncertainty into inevitability. Disagreement also has to survive. Averaging three scientists into one clean rationale reproduces exactly the information loss we are trying to fix. But there is an obvious failure mode here. Once decisions are logged and scored, people log for the record post-facto. Anyone who has watched an electronic lab notebook fill up with retrospective tidying knows how quickly a research tool becomes a compliance exercise. A record that costs a scientist ten minutes of honest reflection per decision is worth far more than one that costs an hour of performance. A publication initiative called Registered Reports offers a useful starting point. Here, researchers submit their rationale, methods and analysis plan for peer review before the data exist, and the journal commits in principle to publish if they follow the approved plan. Nature has now expanded the format across every field it covers. This approach to paper writing timestamps intent before the outcome, but it freezes one plan. A useful search history must also preserve how the plan changed, which alternatives were rejected, when, and on what evidence. Subscribe Run the runner-up Autonomous labs show what happens when decisions and outcomes are connected in a loop. Liverpool’s mobile robotic chemist ran 688 experiments over eight days in a ten-variable formulation space. A batched Bayesian search used each result to choose the next experiments and found photocatalyst mixtures six times more active than the starting formulations. Every completed experiment changed what the system did next. The robot optimized within a goal and search space that humans had already chosen. It did not decide which scientific question mattered. So what this work demonstrates is narrower, but still useful: a search history pays off when the options return standardized outcomes quickly. Open-ended biology is harder because branches can take weeks and the discarded alternatives may never be run. Some exploration therefore has to be bought. Where two branches are plausible and the stakes justify the cost, a lab should sometimes run the runner-up. Otherwise the record captures what today’s scientists usually chose and stays silent on what they systematically overlooked. A funder could ring-fence a small fraction of a grant for the branch not taken, provided that the decision record and the outcome are both deposited. That costs money, but so does having every lab rediscover the same abandoned path. Test whether taste transfers When it comes time to evaluate if our new AI system exhibits taste, the comparison should run across laboratories and fields. At selected decision points in a live research campaign, we should freeze the six fields: the evidence, the candidates, expected results, confidence, costs and the choice made. Experts then rank the options before the outcomes are known, and those outcomes are checked independently later. We could give models one of two diets - papers alone, or papers plus decision histories - and ask them to rank the next experiment on unfamiliar projects. Then, we score information gained per dollar and week, calibration, and how quickly weak branches are abandoned. If histories improve choices only inside the lab that produced them, they amount to useful organizational memory. If they improve choices on unfamiliar problems elsewhere, Alasdair’s group will have captured something science has never managed to write down: a transferable record of taste. Until that happens, a scientific search history is a promising record, not yet a training set.
Claude’s Chrome side panel is now a full Cowork session
Claude is turning its Chrome side panel into a full Cowork session, allowing browser-based tasks and conversations to follow users...