Trump, Johnson, and Tech CEOs Set for September 29 AI Meeting

President Donald Trump will sit down with House Speaker Mike Johnson and chief executives from leading technology companies on September 29 for discussions on artificial intelligence, according to recent reports. Yet with only days left before the session, neither a full guest list nor a detailed agenda has been made public. The reports confirm the meeting is planned, but they do not identify which executives have accepted invitations, and no major frontier lab or large-cap technology firm has confirmed attendance so far. Any confirmations or agenda details that surface before September 29 would offer the first concrete read on the session's direction.

The gathering arrives at a charged moment. Throughout the month, the AI safety debate has gathered momentum: leading developers have issued public warnings about the technology's risks even while shipping new models, while other prominent industry figures have openly challenged calls to slow down.

The Trump administration has already staked out a position on liability. On Monday, Treasury Secretary Scott Bessent said AI developers should take responsibility for their systems rather than expect a federal liability shield — a stance that closes the door on the protective framework some lab executives were suspected of wanting. The remarks set the administration's opening position ahead of the talks and frame liability as a live question rather than a settled one.

What the September 29 meeting will actually tackle remains unsettled: binding regulation, industry-led safeguards, United States competitiveness against rival technology powers, or a combination of all three. Whatever policy perimeter emerges would extend well beyond the labs themselves, spanning the enterprise-technology stack from software platforms such as ServiceNow to networking names like Cisco — none of which has confirmed participation. The lineup also spans the policy process itself: Trump heads the executive branch, and binding regulation, if chosen, would run through Congress, where Johnson leads the House.

For crypto markets, the connection is indirect but material. When the White House convenes the largest technology companies to set the terms for an entire industry, it is defining a regulatory posture that risk assets trade on. Bitcoin (BTC), changing hands near $83,400 at the latest snapshot, is the large-cap most sensitive to a sweeping Washington signal on technology policy — which makes the session a bellwether for digital assets even with an agenda centered on AI.

Labs Keep Shipping Past Slowdown Calls

The road to September 29 runs through a split inside the AI industry itself. Earlier this month, Anthropic researcher Jacob Coxon resigned, alleging that Anthropic and OpenAI were not acting responsibly. His colleague Evan Hubinger went further, putting the probability that AI ends humanity within a decade at more than 10%.

Days after the resignation, Anthropic CEO Dario Amodei published an essay urging the industry to slow the pace at which AI capabilities improve. OpenAI's Sam Altman and SpaceX's Elon Musk backed the call. The essay made clear that pacing would not mean halting training altogether — and every lab that endorsed the idea has released a model since.

Not everyone accepted the slowdown framing. Palo Alto Networks CEO Nikesh Arora questioned the motive, describing the push as a stealthy commercial strategy in a post on X, and later told a television interview he believed the labs wanted sympathy from regulators and a route around liability. Arora's core argument is that pacing is unrealistic: some developers will keep building at the frontier regardless, and those frontier teams are the most responsible operators, so Washington should not try to stop them.

"I think more than likely that some people will jump the gun, and you'll have still people developing at the frontier, which means we shouldn't try and stop the frontiers because they're the most responsible people," Arora said. He also dismissed the 10% extinction estimate as highly unlikely.

The release record backs his read of industry behavior. Anthropic shipped Claude Opus 5.5 on September 22, its first model after the slowdown call, with outside evaluators including METR testing it beforehand. OpenAI released GPT-6 Sol and Luna the same day, following flagship GPT-6 Astra on September 4. Google rolled out Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking on September 15, while DeepMind's Koray Kavukcuoglu confirmed that Gemini 4 has entered post-training, the refinement phase that follows a model's main training run.

For crypto observers, the fight is familiar ground — a young industry negotiating its own liability perimeter before regulators impose one.

Risk Appetite Intact Into the Summit

The through-line is liability: who answers when frontier systems fail. It is the same accountability question crypto settled for itself through pseudonymous wallet addresses — a debate that Bitcoin maximalism long dismissed as tech-industry theater now lands on the White House calendar.

COINOTAG's aggregate market data reads firmly risk-on: the Fear & Greed Index, a composite gauge of crypto market sentiment, sits at 71/100 (Greed), and Bitcoin holds 67.3% of the tracked $2.49 trillion market cap heading into the session. A Greed-territory reading signals optimism rather than defensiveness going into September 29.