Amodei: open weights are 'nowhere near a sufficient solution'

Share
Amodei: open weights are 'nowhere near a sufficient solution'

Anthropic's CEO went where he almost never goes — a long X thread — to make the case for regulation in the middle of the industry's loudest argument about open weights.

Dario Amodei took to X Saturday to defend AI regulation and mandatory frontier-model testing, arguing that open-weight models are "nowhere near a sufficient solution" to AI's concentration problem. The exchange began when a16z's Gavin Baker and Anthropic co-founder Sholto Douglas sparred over claims about Amodei's positions that, in Baker's words, "multiple very serious people in Silicon Valley" believe to be true. Amodei's two-part reply, viewed roughly 490,000 times, rejected what he called the "false choice" between concentrating AI "in the hands of a chosen few companies and politicians via regulation" and distributing it widely. He pushed back on the Silicon Valley shorthand that regulation equals regulatory capture, arguing that fair institutional processes can decentralize power — his analogy was the court system versus mob justice — and that Anthropic's own proposals (California's SB 53, the CAISI and White House testing process, the "Pacing the Frontier" letter) are designed to slow frontier labs while advantaging smaller challengers, "including open-weights!"

The core claim is the interesting one. Amodei argues AI is structurally a power-concentrating technology because of scaling laws, and that open weights only shift the concentration to whoever holds the most compute and chips — "roughly the frontier labs plus maybe hardware providers." That is a direct answer to the open-source camp's central article of faith, and it lands the same week Alibaba put a frontier-adjacent Qwen3.8-27B on home GPUs under an Apache license — Qwen open-sources Qwen3.8-27B that runs on consumer GPUs — the exact challenger dynamic he says regulation should protect. He also endorsed the Trump administration's reported plan for pre-deployment testing of frontier models and of open-weight models as they approach the frontier, plus Demis Hassabis's idea of a FINRA-style regulator, while dismissing the charge that his messaging is disproportionately negative. Elon Musk, Steven Sinofsky, and Brad Gerstner all weighed in, and Techmeme's summary flags his warning of a "crisis of trust" in AI as a through-line.

The significance is partly substance, partly venue. Amodei rarely engages on social media, and his Saturday thread is the most direct public statement of his regulatory philosophy since his June policy essay — made at a moment when the industry is splitting into open-weights and testing camps and Anthropic is heading toward a reported IPO. He is staking out ground that annoys both sides: open-weights advocates get "not sufficient," while the deregulation crowd gets an explicit defense of pre-deployment testing. Whether that nuance survives contact with actual legislation is the open question — but it is now on the record, in his own words, in front of half a million views.

What to watch: whether the administration's pre-deployment testing rules arrive as reported, and how open-weight labs respond to being folded into the testing regime.

Is Amodei right that open weights just move the concentration problem to whoever owns the chips? Write to us via our contact page.

Source: Dario Amodei on X

Get SiliconNoon in your inbox

One email a day with the AI stories that matter — what happened, the numbers, and why it matters. No spam, unsubscribe in one click.

Read more

A local open-weights model built a working license bypass in 30 minutes

A local open-weights model built a working license bypass in 30 minutes

XDA's lead technical editor handed Qwen 3.8 27B — an open-weights model small enough to run on consumer-class hardware — a task he assumed needed a frontier model: reverse-engineering a commercial app's license check. It refused the initial jailbreak, then audited the authentication scheme, recovered a deliberately obscured cryptographic key, caught and corrected its own mistakes, and produced a working bypass — all in about half an hour, entirely offline. The test ran on a single Lenovo…

A chatbot almost sent US troops onto a Chinese ship

A chatbot almost sent US troops onto a Chinese ship

The AI risk debate has spent the week on model weights and pacing pledges. Today it produced something the argument kept promising and never showed: a real near-miss where an analyst's chatbot, not a model's inner life, put an armed boarding party on a launch path. A US military intelligence report that claimed a Chinese ship in the Middle East was carrying components of a nuclear weapons program was built with the help of a chatbot — and was "entirely false," according to CNN, which spoke to…

Bessent and He put open-weight models on the AI agenda

Bessent and He put open-weight models on the AI agenda

Washington and Beijing sat down in New York on Sunday to write the AI chapter of this week's Trump–Xi summit — and for the first time the open-weight question is inside the room, next to the tariff clock. Treasury Secretary Scott Bessent and Vice Premier He Lifeng opened a day of talks at JPMorgan's Manhattan headquarters on Sunday, aiming to tee up agreements on AI, tariffs and critical minerals for the Trump–Xi summit on September 24.

Zhipu open-sources ZCode — but only the version after the fix

Zhipu open-sources ZCode — but only the version after the fix

Zhipu did the thing it promised and published its coding agent's source. What it published starts after the bug. Zhipu (2513.HK) pushed the full source of ZCode to a public repository on Sunday night, nine days after a developer showed the agent was packing entire workspaces and Git history into an encrypted archive only Zhipu could open. The code lives under the company's zai-org organisation, carries an Apache 2.0 licence, and declares version 3.14.0 in its package manifest.

GPT-6 safety report: fewer refusals, more regressions

GPT-6 safety report: fewer refusals, more regressions

GPT-6 reaches ChatGPT's free tier today, an open-source agent just found its price tag, and world models drew a high-profile new contestant. GPT-6 is rolling out to free ChatGPT users today — and OpenAI's 24-page deployment safety report shows exactly what the model traded to become chattier. Plus, Pro, Business and Enterprise tiers started getting GPT-6 Sol on October 7; free and Go users get GPT-6 Luna from October 8, replacing the GPT-5.6 line in ChatGPT's main chat (Work and Codex models…