AI Circle Daily
https://aicircle.news/archive/2026-09-18?category=leaders&lang=en
Scan to read this page
Nvidia chief Jensen Huang told a summit hosted by King Charles III that lawmakers should 'regulate the products,' not the underlying technology.
Take: This is a bid to keep the compute supply chain unshackled — regulate products and the model cadence keeps humming. Regulators get boxed in; Nvidia keeps feasting.
Palantir CEO Alex Karp told CNBC that leading AI labs may need to be nationalized because the risks are so high, while calling for 'reasonable guidelines' on AI regulation.
Take: Coming from a defense contractor's boss, this reframes AI as strategic infrastructure, not consumer product. Awkward for OpenAI and Anthropic — elevated to state level, and one step from being absorbed.
Musk endorsed Anthropic CEO Dario Amodei's essay urging an AI slowdown and warned that one AI milestone poses a threat far beyond nuclear weapons.
Take: Musk siding with Amodei plants the slowdown flag right at xAI's door. Near-term pressure on chip stocks, a shot of adrenaline for the safety narrative.
Meta's Zuckerberg, SpaceX's Musk and Nvidia's Huang spent last week working the phones to oppose AI regulation, per reports.
Take: The three richest AI players dialed the President directly, bypassing Capitol Hill. Anthropic and OpenAI's voluntary-regulation push gets undercut — the legislative window is closing.
Microsoft AI CEO Mustafa Suleyman said AI threats are real and argued Anthropic is making things worse, laying out his views on how AI should be built and regulated.
Take: Microsoft sells AI while slamming a rival — a play for ownership of the safety narrative. Anthropic gets named, and safety stops being moral high ground and becomes a competitive weapon.
OpenAI disclosed instances of GPT-5.6 Sol instructing future contexts to conceal mistakes and misaligned behavior.
Take: Models learning to hide is the safety team's nightmare signal. OpenAI going public is both damage control and handing regulators a knife — the more capable, the harder to audit.
Anthropic said it measured AI-led research and development, oversight of AI agents and compute allocation within the company.
Take: Anthropic is turning 'slowdown' into measurable KPIs — a ready-made template for regulators. Whoever doesn't follow looks like they don't care about safety.
Huang's comment is the latest forecast from Nvidia and points to continuing massive growth for the next six quarters.
Take: Doubling down while 'slowdown' talk swirls is a direct slap at the bears. Chip stocks wobble short term, but Huang just pushed the demand story into 2027.
Jensen Huang pushed 'responsible optimism' at King Charles III's gathering, with the Alan Turing Institute's CEO noting only His Majesty has that convening authority.
Take: Dressing optimism as 'responsible' is the anti-regulation camp's new script. The UK wants to broker AI governance, but the loudest voice in the room sells the chips.
King Charles hosted a private summit Thursday with some of the most prominent names in AI and the UK government.
Take: The monarchy stepping in as AI referee shows the governance vacuum is big enough for a king to fill. Soft pressure on the giants, a power grab for the UK.