DC to Silicon Valley: Shut the Hell Up
DC to Silicon Valley: Shut the Hell Up
Washington has always kept Silicon Valley at arm's length. The cultural gap is real. The policy vocabulary is different. The incentive structures point in opposite directions. But the current friction between lawmakers and AI executives has moved past cultural disconnect into something more consequential: active political liability.
That is the message coming out of both parties right now, according to Axios, which spoke with lawmakers and White House advisers for a report published this week. Stop cycling between utopian prosperity and existential catastrophe in the same breath. Stop making a complicated regulatory debate into an impossible moral panic. Build your products. Do it safely. And stop making it worse every time you open your mouth.
The Self-Inflicted Wound
The core complaint from Washington is not that AI is dangerous. It is that AI executives are turning a policy debate that was already difficult to govern into something nearly impossible to act on rationally.
The standard AI CEO public appearance has become a rhetorical whiplash machine. In one breath, these leaders describe AI as the greatest prosperity-generating technology in human history. In the next, they warn that it could end civilization. Both pitches land with complete conviction. Neither one is easy to build legislation around.
That pattern is not strategic calculation. It does not manage regulatory risk on two fronts. It just creates instability at exactly the moment Washington is trying to figure out what a governing framework should look like.
Anthropic's Bad Month
Nobody is having a rougher stretch in Washington right now than Anthropic.
Dario Amodei attended a late September gathering with President Trump and reportedly did not disclose safety incidents involving Anthropic models that affected government websites. U.S. officials were notified Thursday, October 8. A blog post followed the very next evening. Washington noticed the 24-hour turnaround.
Then Anthropic introduced an AI safety bill that included federal preemption of state AI laws — a position the company has broadly opposed unless strict federal regulations are established first. The bill also defined a catastrophic AI incident as one causing 50 or more deaths. The political backlash was immediate. A senior Anthropic policy official has since walked back parts of the bill.
The result is a company that looks, at minimum, disorganized — and, at worst, like it is trying to manage its narrative quietly ahead of a significant corporate milestone. Neither read is good for a company whose entire brand proposition is that it takes safety more seriously than anyone else.
You cannot build public trust on a brand of safety while treating major policy disclosures like Friday evening news dumps.
OpenAI Is Not Doing Any Better
OpenAI has its own Washington problem. The company's legislative strategy has drawn skepticism on Capitol Hill and in state legislatures. New York assemblymember Alex Bores — a computer scientist — argued publicly that OpenAI's policy proposals are deliberately vague, designed primarily to block state-level regulation without offering a credible federal alternative in its place.
A developing FTC investigation targeting both companies adds a layer of institutional pressure that makes every stumbling public statement a potential exhibit.
Meanwhile, the political landscape keeps shifting against the industry. Recent polling shows a majority of Americans now view AI as bad for society. That is not a marginal result. It means the industry has failed at basic public persuasion while simultaneously alienating the lawmakers it needs to build any workable policy environment.
A Rare Bipartisan Agreement
The "shut up and build" message from Washington carries unusual weight because it is genuinely bipartisan. Republican and Democratic lawmakers are saying the same thing in different accents.
The Republican critique focuses on cultural disconnect. AI leaders are operating in a vacuum, out of touch with ordinary Americans, cycling between doom and boosterism without landing anywhere that makes sense to voters. The Democratic critique focuses on power. AI money has cozied up to the administration. Workers feel threatened. The industry has not made a convincing case that its gains should be shared broadly.
Those are different political arguments. They converge on the same conclusion: the AI industry's public communications strategy is not working, and it is making everyone else's job harder.
Republican or Democrat, the message is the same. Stop making this harder than it already is.
The Lesson Washington Is Teaching
There is an obvious irony in Washington telling anyone to communicate less. But the frustration here is real and substantive.
The AI industry had a genuine opportunity to lead its own narrative. It could have framed a clear public case for shared prosperity. It could have explained the technology in terms that land with ordinary people. It could have built the kind of political goodwill that makes sensible regulation possible.
Instead, it handed the microphone to executives who alternate between selling a miracle cure and warning about the apocalypse. The result is a skeptical public, a frustrated Congress, and a White House that keeps waking up on Monday morning to a new AI headline that nobody needed.
The advice coming out of Washington is not complicated. Build better products. Communicate with some basic consistency. Stop treating major policy disclosures like narrative management exercises. The AI industry knows how to ship software. It has not yet figured out how to be a responsible public actor.
Until it does, the message from Washington will keep being the same one. Just stated less politely each time.
Aaron Rose is the founder and publisher of Tech Reader Magazine, covering system architecture, cloud platforms, and AI policy.