Concerns are rising among shareholders about Dario Amodei, the CEO of Anthropic, whose unpredictable demeanor and public outbursts are seen as potentially harmful to the $380 billion AI company. Recently, Amodei, known for creating the Claude chatbot, made headlines by criticizing President Trump and OpenAI’s Sam Altman in a contentious internal memo. This coincided with the Pentagon’s decision to blacklist Anthropic because the company declined to remove safeguards on its AI’s military applications.
The memo, which leaked to the media soon after it was sent, accused the Pentagon of targeting Anthropic due to the company’s unwillingness to “praise Trump in a dictator-style way” while suggesting that Altman was spreading “blatant lies” regarding shared concerns for AI safety. A current shareholder, who preferred to remain anonymous, described Amodei’s aggressive remarks as part of a “deeply disturbing” trend for a CEO representing high-profile investors like Amazon and Google.
“You’re a CEO who’s raised billions. You can’t just rant and expect all shareholders to agree with you,” the shareholder remarked, noting the unusual severity of Amodei’s opinions, particularly given the Pentagon’s central role in AI technology development.
Amodei has often found himself at the center of controversy. He has previously suggested that advancements in AI could lead to a 20% increase in unemployment and compared the decision to sell advanced AI chips to China to selling nuclear weapons to North Korea. Shareholders pointed out that despite efforts from others within the company, like co-founder Daniela Amodei and policy director Jack Clark, to temper his statements, he still tends to speak impulsively.
“They’re trying to help him, but it seems like he can’t control his emotions,” the shareholder added. However, some officials close to Anthropic defended Amodei, highlighting the company’s rapid growth and support from various investors amid the Pentagon clash.
Amodei’s animosity towards Altman began even before he left OpenAI in 2021 to establish Anthropic, after what he considered reckless decisions on AI safety. Their rivalry was made evident in a tense moment during a tech conference in India, where they both avoided holding hands for a group photo in front of an enthusiastic crowd.
In the latest development, some insiders claim Amodei’s fixation on OpenAI could be detrimental to Anthropic’s overall mission. He has reportedly expressed contempt for OpenAI’s team, questioning their loyalty to Altman. “It’s evident this is personal for him, but he rationalizes it through a belief that Altman won’t prioritize safety,” a shareholder mentioned.
Meeting within Anthropic often feature Amodei leading discussions about various subjects, indicating he still has a strong base of support within the company. Recently, he expressed regret for his memo about the Pentagon, stating that it wasn’t in Anthropic’s interest to escalate tensions.
A shareholder remarked, “Any reasonable CEO would know the risk of writing something like that,” attributing Amodei’s actions to his inflated self-regard and the encouragement he receives from those around him.
Anthropic is facing legal issues with the Department of Defense over its classification as a “supply chain risk,” arguing the government’s actions are unwarranted. A senior official from the Pentagon suggested this designation was necessary due to the differing ideological foundations of Anthropic’s AI compared to the Pentagon’s needs. A federal judge recently issued a preliminary injunction that temporarily blocks this ban.
Despite the surrounding chaos, the company’s pursuit of defense contracts appears contradictory to its AI safety philosophy. Nevertheless, some insiders believe Amodei is eager to be involved in critical discussions on advanced AI use in military contexts.
Critics have suggested that Anthropic’s emphasis on safety might just be a tactic to gain regulatory advantages over competitors. While many investors have shown support for Amodei amid the Pentagon controversy, some have privately expressed concerns about his combative rhetoric, especially as tensions with the Trump administration escalated.
As the deadline for an agreement with the Pentagon loomed, Amodei declared that Anthropic could not comply with their requests “in good conscience.” A source framed the situation as a matter of personal ego and diplomacy.
