Mike Archer

Explorations into AI

[Mike Archer @MikeArcherAI]

This all seems great in theory, but the solution is a healthy competitive market in which models can check each other.

The best and only regulation should be to ensure a non-monopoly. We want a diverse, competitive ecosystem in which AIs compete for outcomes.
@elonmusk

[Dario Amodei @DarioAmodei]

We Must Pace the Frontier: I’ve written a new essay on why the AI industry should slow down, with a three-part plan for doing so.

Anthropic is unilaterally committing to the first of these steps. We’ll provide third-party evaluators with permanent, employee-level access to our systems, so that they can verify adherence to our safety measures, report on incidents, and assess models’ alignment during training.

You can read the full post here: https://t.co/OGyPb7yaYt

[Mike Archer @MikeArcherAI]

@jakeglmn @grok are his claims true?

[Sayer Ji @sayerjigmi]

🚨 UPDATE: In May, the Gates Foundation and Anthropic launched a $200M partnership covering global health and life sciences.

On Sept 17, Anthropic loosened its safeguards for biology work.

On Sept 18, Reuters revealed it quietly built a physical biology lab. It won't say what it's for.

Three dates. Four months. One company.

MAY 14, 2026 — Anthropic and the Gates Foundation announce a four-year, $200 million partnership. The Gates Foundation grants funds. Anthropic contributes Claude usage credits and staff support. The stated scope: global health, life sciences, education, economic mobility.

SEPT 17, 2026 — Anthropic introduces the Life Sciences Verification Program. Verified life-science professionals get access to its Mythos, Opus, and Sonnet models under what Anthropic's own announcement calls "a refined set of safeguards MORE PERMISSIVE for biology-related work." It is the first time Mythos has been opened this way.

SEPT 18, 2026 — Reuters reports Anthropic has quietly built a wet lab — a facility for physical biological experiments. Its head of life sciences, Eric Kauderer-Abrams, confirms it on the record: "We believe that to do biology, the final test is still and will be for a while in real lab work."

Then a spokesperson told Reuters the lab is "not for drug discovery specifically" — and declined to elaborate.

Read that again. The company confirmed the lab exists, then declined to say what it does, offering only what it isn't.

One person familiar with the work told Reuters that Anthropic wants Claude to "direct robotic units to carry out science experiments with limited human intervention." Anthropic's spokesperson says human oversight remains essential for safety. Both statements are in the same article.

Here is why this matters.

The entire regulatory case being built right now — the Sanders bill, the UK superintelligence bill, the closed Senate briefings — rests on the premise that frontier AI could help someone engineer a pathogen. That is the load-bearing argument. It is why safeguards on biology exist in the first place.

On September 17, the company that has done more than any other to advance that argument made its biology safeguards more permissive.

On September 18, we learned it had built a lab to do physical biology.

Safeguards for you. A wet lab for them.

This is the pattern I documented in my investigation of the Gates AI paradox: the same capital advances the capability and funds the alarm about the capability, and the rules that follow land on everyone except the people who wrote them.

I did not expect the next data point to arrive within twenty-four hours.

WHAT I AM NOT CLAIMING: I am not claiming the Gates Foundation funded this laboratory. The May partnership commits funds to programs implemented with outside partners, not to Anthropic's internal facilities. I am not claiming Anthropic is developing biological weapons, or that its lab is unlawful, or that anyone coordinated these three announcements. Every fact above is on the record and linked below.

What I am claiming is that a company whose public safety argument centers on biological risk quietly built a biology lab, loosened its biology restrictions the day before that became public, and will not say what the lab is for.

That is not a conspiracy theory. That is a press release, an announcement page, and a Reuters interview.

My investigation into who funds both sides of the AI safety debate below.

[Brian Roemmele @BrianRoemmele]

A warning from someone who knows, he helped build the AI act in Europe and regrets it:

"Europe actually passed this AI Act some years ago that puts draconian restrictions on AI.... Europe is now out of the AI race. I hope we're not gonna let the same thing happen in America."—Professor @pmddomingos

[Peter J. Hasson @peterjhasson]

The Wall Street Journal just confirmed that Jacob Coxon lied here. He worked with multiple "AI safety" execs ahead of his resignation.

[Mike Archer @MikeArcherAI]

Wait till you see what I uncovered. Video releases ~9/25

[Peter J. Hasson @peterjhasson]

The Wall Street Journal just confirmed that Jacob Coxon lied here. He worked with multiple "AI safety" execs ahead of his resignation.

[Mike Archer @MikeArcherAI]

@pmddomingos @pmddomingos I went even deeper down the rabbit hole, found an effort to influence hundreds of millions to billions of people (know it sounds ridiculou, but have receipts). Should have a video in a week or so.

[Wall St Engine @wallstengine]

Mark Zuckerberg, Elon Musk and Jensen Huang reportedly pushed Trump to drop a proposed AI regulator backed by Google DeepMind chief Demis Hassabis.

The plan would have created an industry-funded body to set standards and oversee AI companies, similar to FINRA in finance.

[David Sacks @DavidSacks]

Joe Lonsdale: “These companies should be held extremely liable for any damage they do. We shouldn't be putting regulators in. They're just going to get captured... The right should be against regulation, but for liability to hold companies accountable.”

[Naval @naval]

The best way to pace the frontier is to hold the labs fully liable for the behavior of their models.

[Bill Gurley @bgurley]

I'm not of fan of regulation for two reasons - 1) its hard to get right, and 2) its highly susceptible to regulatory capture. That said, if we insist on doing it, we should try to get it correct.

As "safety" seems is the primary goal of this AI regulation exercise, I wanted to share that I recently went deep on both the 737 Max (2 crashes) and Fukushima accidents. In both cases, the investigations led to a conclusion that "regulatory capture" was the EXACT failure that led to both safety incidents. Boeing/FAA and Tepco/NISA.

As a result, with safety as the #1 goal, the regulator MUST BE someone with authority and accountability "over" the regulated company. The idea that it could be a friendly third party with lots of cross-relationships is structurally broken idea. Ask Mr. Kiyoshi Kurokawa.

If we insist on doing this the regulator (1) needs to be 100% independent, 2) the "tests" and "procedures" need to come for the regulator not the company, and (3) the regulator needs to understand their responsibility is to the American people. They need to feel the authority over the regulated company and the responsibility to the nation.

It's imperative. Or it will fail.

[Polymarket @Polymarket]

JUST IN: Chinese robots have begun building other Chinese robots, as a new factory capable of producing 1 humanoid every 10 minutes comes online.

[Chamath Palihapitiya @chamath]

Elon’s idea of peer review is excellent and will incentivize each frontier lab company to invest in test harnesses and auditability.

This is infinitely better than some transnational governance body to decide on humanity’s future.

[ℏεsam @Hesamation]

WOW... Anthropic has a surprisingly dense web of connections to the world’s largest funder of "catastrophic-risk" work (aka AI doom):

> Open Philanthropy → built by Karnofsky + Moskowitz’s Good Ventures
> Good Ventures → funded Jacob Coxon’s scholarship
> Karnofsky → married to Daniela Amodei, Dario's sister and Anthropic's president
> Moskowitz → early investor in Anthropic
> Dario → scientific advisor to Open Phil years before Anthropic
> Karnofsky now works at Anthropic on AI safety strategy
> Open Phil-funded MATS → Evan Hubinger → now Anthropic alignment-science lead
> Hubinger backed Coxon: "We really do earnestly believe AI could kill all humans!"
> Ethan Perez → Open Phil fellow + ~$1M Open Phil research grant → Anthropic Head of Alignment
> also backed Coxon: "100% agree with [Coxon] that AI poses serious risks to society, and I'm glad he's speaking out!"

is this kind of a dense network normal for a company?

[Mike Benz @MikeBenzCyber]

which worked famously well when we did this with coronavirus research

[The All-In Podcast @theallinpod]

America's Next 250: Remember Who We Are

All-In Summit's Opening Video:

[jack @jack]

https://t.co/qAUP2dNJsi

[Ben Pouladian @benitoz]

Holy crap POTUS just phoned in @JensenHuang live on stage at All In Summit

We will not lose the ai race! And whatever Dario said this weekend won’t stop our progress

This made my morning!

$NVDA

[Mike Archer @MikeArcherAI]

@JDVance And for those major skeptics that think I changed the title, watch the first 5s of the video and you'll see I use the exact language "begging" and "trojan horse".

[Mike Archer @MikeArcherAI]

Is @JDVance watching my AI videos? haha. Crazy that we're all converging on the same hypotheses and language independently.

[Trump War Room @TrumpWarRoom]

VP @JDVance: "Personally, I feel a little bit weird about the fact that you have so many frontier AI tech companies kind of coming to the government and begging the government to regulate them. It feels a little bit to me like a bit of a Trojan horse."

[Mike Archer @MikeArcherAI]

@TrumpWarRoom @JDVance Either @JDVance is watching my videos or we're converging on the same hypotheses and language, CRAZY! ("Begging and Trojan Horse")

https://t.co/ASbWVCoLeo

https://t.co/WVqljdj6o5