President Trump hosted a White House luncheon on Tuesday and walked out with a signed document called the White House Accord on Super Intelligence. The agreement, described as “morally binding,” asks tech companies to regulate themselves.
The accord, discussed on this week’s Uncanny Valley podcast, brings together Sundar Pichai, Elon Musk, Dario Amodei, Mark Zuckerberg, Jensen Huang, and OpenAI President Greg Brockman. OpenAI CEO Sam Altman was notably absent. The companies promise to put their models through four layers of controls and audits, including internal evaluations, outside audits, and board reviews.
The Lunch That Spawned an Accord
Trump invited the executives to the White House for a sit-down meeting. The group includes leaders from major tech firms, smaller companies, and AI labs. The gathering was billed as a discussion of AI safety, but the final product is a statement of intent.
The accord’s name, “Super Intelligence,” is a government term for AI. The document asks companies to apply four layers of controls and audits to their models. Those layers include internal evaluations, outside audits, and board reviews.
Trump praised the arrangement during the event. The hosts on Uncanny Valley noted that the takeaway message from the White House was simple: “We got this. We take it seriously. Don’t worry about it.”
Dario Amodei’s Awkward Moment
One moment from the luncheon has gathered attention online. Trump asked Amodei to step forward and speak, praising him and saying “Whatever he says is OK.” The hosts note that Amodei did not appear eager to be in the front of the room.
The exchange is a small detail, but it captures the tone of the whole affair. A leader praised by the president is not necessarily a willing participant.
OpenAI’s Developer Day Announcement
The same week, OpenAI held its Developer Day. CEO Sam Altman introduced the company’s new always-on AI agents for regular users, which the hosts call “AI agents for normies, otherwise known as Dots.” The agents are designed to work alongside humans in everyday settings.
Employers Push AI Agents Into the Workplace
As AI agents grow more common, employers are pushing to have them work alongside human employees. The hosts note that the process is underway and that the chances of bots joining the workforce are high.
The question of how well humans and AI agents will collaborate is open. The hosts frame the situation as a matter of what the chances are that we’re all going to have to welcome some bot colleagues very soon.
Extremists Running for Office
The midterm elections are closing in, and the podcast takes a look at some of the extreme candidates running for office. The hosts point out that it’s not just political candidates who feel emboldened to say whatever they want. People who are currently or have been recently working for the government are also speaking out.
One example cited is a retired army colonel turned Pentagon influencer who posted praise for dictators and has a plan to strip the vote from anyone without kids. The hosts describe this as part of a pattern of emboldened speech among government figures.
What the Accord Actually Does
The White House Accord is a voluntary agreement. The companies sign on, but the hosts on Uncanny Valley are skeptical of the arrangement.
The hosts note that the word “voluntary” is the keyword throughout. They wonder aloud whether none of this is real, calling it all fake.
The Skepticism Behind the Accord
The hosts on Uncanny Valley are clearly doubtful that self-regulation will work. The voluntary nature of the accord means that companies are free to ignore it.
The skepticism is rooted in the history of voluntary agreements. Companies have incentives to cut corners, and there is no external force to hold them accountable.
The hosts’ doubt is the central tension of the episode. They present the accord as a gesture without substance, a pinky swear that carries no weight.
Key Facts Box
- Event: White House luncheon on Tuesday
- Agreement: White House Accord on Super Intelligence
- Signatories: Sundar Pichai, Elon Musk, Dario Amodei, Mark Zuckerberg, Jensen Huang, Greg Brockman
- Not present: Sam Altman, OpenAI CEO
- Layers of controls: Internal evaluations, outside audits, board reviews
- Status: Voluntary
The White House Accord is a statement of intent dressed up as regulation. It asks companies to do the right thing and trusts them to follow through. The hosts on Uncanny Valley are not convinced that trust will hold.
The episode closes with a note about the live audience recording of the podcast on Thursday, October 8, at 7 pm at The Commons in San Francisco. Tickets are available on kqed.org.
For listeners, the message is clear: the AI future is coming fast, and the rules are still being invented.
Source material: “Trump’s ‘Morally Binding’ AI ‘Accord,’ the Rise of AI Agents, and Extremists on the Ballot,” WIRED.
Get the Notebook.
The day's best stories and every fresh verdict, in plain English, in your inbox by seven. One email a day, no more.