AI Circle Daily
https://aicircle.news/archive/2026-09-20?category=leaders&lang=en
Scan to read this page
Nvidia CEO Jensen Huang dismissed predictions that AI will end humanity by 2030, asserting the technology poses no extinction risk and advocating for continued rapid development.
Take: The shovel-seller's real fear isn't doomsday — it's a regulatory brake. Huang is effectively underwriting the capex and compute arms race.
Speaking virtually at the G20 Innovation Ministerial, Elon Musk predicted humanoid robots could number over one billion within 10 years and collectively outproduce humanity.
Take: Musk is shifting the narrative from EVs to humanoids to reprice Tesla. Buy the story and you also buy his safety talk as a shield against regulation.
California Gov. Gavin Newsom issued an executive order to rein in AI before it's too late, joining other 2028 Democratic presidential hopefuls in calling for a more aggressive approach to AI fears.
Take: State-level rules landing before federal ones front-loads compliance costs onto Silicon Valley. Build models in California, take the hit first.
Virginia Gov. Abigail Spanberger ordered an AI task force and moved to restrain data center development, banning executive branch actions and giving local communities more say in approvals.
Take: The data center capital braking its own buildout is the clearest signal yet: AI's bottleneck is shifting from chips to power, land and local politics.
Microsoft's Mustafa Suleyman told CNBC's Squawk Box that OpenAI's latest incidents of concerning model behavior disclosed this week represent a serious situation.
Take: Microsoft, the biggest backer, publicly framing this as serious puts a question mark on OpenAI's safety record — aimed at regulators and cloud customers alike.
Anthropic selected Accenture as its first embedded evaluator to help implement CEO Dario Amodei's slowdown proposal, amid intense scrutiny of Anthropic and OpenAI over potential catastrophic AI harm.
Take: Outsourcing safety evals to a consulting giant is Amodei's play to self-regulate before being regulated. Whether Accenture is truly independent will be watched hard.
A coalition of over 100 AI experts urged independence and transparency from Anthropic, OpenAI and other foundation model labs in conducting evaluations.
Take: The letter hits the soft spot for Amodei and Altman: choosing your own evaluator is judging your own case. If independent evals become standard, the labs' narrative costs spike.
Trump said it's time to rebrand AI with a new name and announced he's creating an AI Force, while claiming without evidence that the AI backlash is a Democratic hoax.
Take: Politicizing AI and weaponizing naming rights adds another layer of policy noise. Whoever has Washington's ear gets the signal first.
In a wild week, Musk agreed with the CEOs of Anthropic and OpenAI on AI safety while fighting regulation, and was contradicted by President Trump and Nvidia's Jensen Huang.
Take: Musk's flip-flopping between safety talk and anti-regulation exposes xAI's bind: wants a seat in the safety club, won't be locked down by rules.
Anthropic is operating a lab that conducts biology experiments, even as its researchers warn AI might kill us all while leaders promise AI is key to curing disease.
Take: Cure and kill in the same breath puts Anthropic in the most contradictory spot. The lab is both an R&D asset and a live ad for its safety narrative.