Anthropic CEO Dario Amodei went public Tuesday to defend his company.
He said he needed to "set the record straight" after "a recent uptick in inaccurate claims about Anthropic's policy stances."
His core message: "Anthropic is built on a simple principle: AI should be a force for human progress, not peril. That means making products that are genuinely useful, speaking honestly about risks and benefits, and working with anyone serious about getting this right."
The statement came after Trump administration officials spent last week attacking the AI company.
It started when Anthropic co-founder Jack Clark shared his thoughts on AI.
Clark expressed what he called "appropriate fears" about AI. He described AI as a powerful, mysterious, "somewhat unpredictable" creature - not a dependable machine that's easily mastered and put to work.
White House AI czar David Sacks pounced.
His response: "Anthropic is running a sophisticated regulatory capture strategy based on fear-mongering. It is principally responsible for the state regulatory frenzy that is damaging the startup ecosystem."
That's a serious charge. Sacks is essentially accusing Anthropic of: • Exaggerating AI risks • Pushing for regulations • Doing so to hurt competitors and entrench themselves
Other Trump officials and AI leaders joined in.
White House senior policy advisor for AI Sriram Krishnan also criticized Anthropic for stoking fears to damage the industry.
California Senator Scott Wiener - who authored AI safety bill SB 53 - defended Anthropic. He called out Trump's "effort to ban states from acting on AI w/o advancing federal protections."
Sacks doubled down. He claimed Anthropic was working with Wiener to "impose the Left's vision of AI regulation."
Groq COO Sunny Madra piled on, saying Anthropic was "causing chaos for the entire industry" by advocating for AI safety measures instead of "unfettered innovation."
This fight exposes a fundamental divide in Silicon Valley and Washington.
One camp (Sacks, Krishnan, Madra): AI should develop with minimal regulation. Let innovation run free. Restrictions hurt startups and hand advantages to big players.
The other camp (Anthropic, Wiener): AI is powerful and unpredictable. Some guardrails are necessary. Being honest about risks isn't fear-mongering - it's responsible.
The "regulatory capture" accusation is particularly loaded.
Regulatory capture happens when companies push for regulations that benefit themselves while hurting competitors. For example, big companies lobby for expensive compliance rules that only they can afford, squeezing out smaller rivals.
Sacks is saying that's Anthropic's game plan.
Amodei pushed back against the characterization.
His statement emphasized "working with anyone serious about getting this right." That's a subtle jab - suggesting critics aren't serious about responsible AI development.
Anthropic has been vocal about AI safety. The company was founded partly on concerns that AI development was moving too fast without enough attention to risks.
But being cautious about AI doesn't mean opposing innovation. Anthropic is actively building AI products while advocating for safety measures.
This plays into larger Trump administration priorities.
The White House has been hostile to AI regulation. Trump and his team see restrictions as obstacles to American AI dominance, especially versus China.
State-level AI bills like California's SB 53 particularly anger the administration. They want federal control over AI policy - or better yet, minimal policy at all.
Anthropic speaking honestly about AI risks complicates that narrative. If AI really is "somewhat unpredictable," maybe some oversight makes sense.
That's why Sacks is hitting so hard. He needs the narrative to be: AI is safe, regulation is unnecessary, critics are just protecting their own interests.
This is more than a corporate spat. It's a fight over AI's future.
Should companies be allowed to develop AI with minimal oversight? Or do we need guardrails given AI's power and unpredictability?
Anthropic's position: Being honest about risks isn't fear-mongering. It's responsible. AI is genuinely powerful and not fully understood yet.
The Trump administration's position: Anthropic is exaggerating dangers to push regulations that benefit themselves and hurt competitors.
For investors and the public, the stakes are real.
If Sacks is right, Anthropic is cynically manipulating policy for competitive advantage. That would be bad corporate behavior worth criticizing.
If Amodei is right, the Trump administration is trying to silence legitimate concerns about AI safety to rush development without proper safeguards. That would be reckless.
The truth probably lies somewhere in the middle. AI does have real risks worth discussing. And companies do sometimes push regulations that benefit them.
But the intensity of this fight shows how high tensions are running. The Trump administration clearly wants full speed ahead on AI with no speed bumps. Companies like Anthropic saying "maybe we should be careful" threatens that agenda.
Amodei's statement won't end this. Expect the attacks to continue. The administration wants AI companies to shut up about risks and focus on building. Companies that won't play along will keep getting hammered.
For Anthropic, the question is whether standing firm on safety principles is worth the political heat. So far, they're not backing down.
