Fiori D'Arancio News

Sam Altman is stuck playing defense. It all started a week ago.

· Business Insider

OpenAI CEO Sam Altman

Kevin Wolf/AP

Visit milkshakeslot.com for more information.

  • Sam Altman and OpenAI face backlash after Pentagon deal raises ethical concerns.
  • Anthropic CEO Dario Amodei rejected Pentagon terms, prompting a heated response from officials.
  • Anthropic's Claude app surged in downloads as users react to OpenAI's Pentagon deal.

Days after triumphantly declaring victory and landing a big deal with the US government, Sam Altman is stuck playing defense.

In an interesting twist, OpenAI's latest debacle is arguably its least technical challenge yet. It also means Altman can't just push out an app update to alleviate the issue.

Altman and OpenAI are now in damage control mode after their decision last week to strike deal with the Pentagon for access to its AI models after their rival Anthropic refused to agree to the government's terms.

It's become a fast-moving story filled with contract negotiations, optics management, and a tipping of the scales in the chatbot wars as the public has watched it all unfold, voting with their downloads and wallets.

So how did we get here? Whether you're playing catch-up or just plain confused at where the drama began in the place, we've got your back.

Here's the play-by-play of OpenAI's rocky last week.

Tuesday, February 24: Anthropic gets an ultimatum

Anthropic CEO and cofounder Dario Amodei

Chance Yeh/Getty Images for HubSpot

Before this story became about OpenAI, it was about Anthropic CEO Dario Amodei, Altman's former colleague and now rival. (If you want to get a sense of just how bitter their feud is, the pair refused to hold hands during a recent photo op with other AI execs).

For days, the Pentagon had threatened Anthropic over its refusal to allow the nation's military unfettered access to its AI systems. As tensions came to a head, Defense Secretary Pete Hegseth summoned Amodei for a face-to-face meeting at the Pentagon.

A Defense Department official later told Business Insider that Hegseth delivered an ultimatum to Amodei: Get on board by 5:01 p.m. Eastern Time on February 27th or get essentially blacklisted by the US government.

Wednesday, February 25: OpenAI begins its own talks

A source familiar with the situation told Business Insider that OpenAI began its talks with the Pentagon on Wednesday.

But at this point, all eyes are on Anthropic and whether it will bend to the Pentagon's demands or risk losing its business — and potentially facing other repercussions.

Thursday, February 26: Anthropic CEO details the AI company's red lines in Pentagon talks

Amodei, a biophysicist by training, is known for his zealous pursuit in the safe development of AI. In fact, it's why he and six other cofounders, including his sister Daniela, said they left OpenAI to start their own company in the first place.

Anthropic wanted to work with the US military, Amodei said, but in two specific instances it wanted safeguards to make sure that its technology was used safely.

Anthropic

Amodei said Anthropic "could not knowingly provide" its AI to be used to guide fully autonomous weapons, because he said such technology could not be reliably deployed in combat. Amodei was also concerned that Anthropic's AI could effectively used to conduct mass domestic surveillance, even if such searches may not violate the letter of the law.

"We cannot in good conscience accede to their request," Amodei said in a statement. He expressed hope that the Pentagon would reconsider its threats about what would happen to Anthropic if it didn't acquiesce.

In response, Emil Michael, the Pentagon's top R&D official, wrote on X, "@DarioAmodei is a liar and has a God-complex."

Friday, February 27: Trump lashes out at Amodei and Anthropic

Amodei didn't have to wait long for a response after rejecting the Pentagon's reported final offer. Just over an hour before the deadline, President Donald Trump unloaded Anthropic, calling the AI startup a "RADICAL LEFT, WOKE COMPANY."

"I am directing EVERY Federal Agency in the United States Government to IMMEDIATELY CEASE all use of Anthropic's technology," Trump wrote on Truth Social. "We don't need it, we don't want it, and will not do business with them again! "

Later on, Hegseth said he intended to formally declare Anthropic to be a supply chain risk to national security, a designation that he said would effectively ban any company that received Defense Department contracts from doing business with the AI startup. (Anthropic responded that such a sweeping designation was "unprecedented" and they would challenge it in court if enacted.)

Altman strikes a deal where Anthropic wouldn't

Late Friday evening, Altman announced on X that OpenAI had struck a deal with the DOD "to deploy our models in their classified network."

He also said he didn't support the idea of the US government labeling Anthropic a supply-chain risk, and that he had made his position clear to officials.

While some OpenAI employees publicly celebrated the deal, backlash within the AI community quickly followed, including from former top OpenAI employees.

"In light of what external lawyers and the Pentagon are saying, OpenAI employees' default assumption here should unfortunately be that OpenAI caved + framed it as not caving, and screwed Anthropic while framing it as helping them," Miles Brundage, OpenAI's former head of policy, wrote on X. (Brundage later added that he shouldn't have used the word "caved.")

People quickly zeroed in on government officials' claims that OpenAI would allow the Department of Defense to use its AI models in "all lawful" purposes, with critics suggesting it left the door open to a use case that Anthropic said was a line it was unwilling to cross.

Saturday, February 28: ChatGPT defectors help boost Claude to No. 1 on the App Store

Anthropic skyrocketed in popularity as OpenAI endured criticism over the deal. Some ChatGPT users decided to vote with their phones and defected to Anthropic's Claude chatbot.

The show of support for Claude quickly translated into downloads.

On Saturday, Claude became the top downloaded app on Apple's US App Store. Weeks earlier, Claude hadn't even been in the top-40, but now Anthropic had blown past competitors like OpenAI's ChatGPT, Google's Gemini, and xAI's Grok.

Pop star Katy Perry threw her support behind Anthropic, sharing a screenshot of signing for the company's $214.99 annual Claude Pro plan.

Altman turns to defense

As criticism mounted, Altman fielded questions from users on X during an informal AMA on Saturday evening, acknowledging that OpenAI's deal "was definitely rushed, and the optics don't look good."

Ultimately, Altman said he wanted OpenAI's agreement "to de-escalate things." He said time will tell whether that was the correct move.

Monday, March 2: Altman says OpenAI amended its government agreement

Following up on President Donald Trump's statement, the Treasury Department announced it would no longer use Anthropic's products.

Altman later announced that OpenAI had amended its agreement to add more specific protections to prevent the Pentagon from using OpenAI's models to conduct mass domestic surveillance.

"It's critical to protect the civil liberties of Americans, and there was so much focus on this, that we wanted to make this point especially clear, including around commercially acquired information. Just like everything we do with iterative deployment, we will continue to learn and refine as we go," he wrote in an internal message that he later shared publicly on X.

Crucially, the language still cites current US law, which Amodei had repeatedly said is outdated and does not adequately address how the government could use AI to effectively spy on Americans, even if no laws are violated, thanks to the capabilities of AI to comb through copious amounts of publicly and commercially available data.

Altman reiterated that OpenAI shouldn't have "rushed" its deal and that the Pentagon shouldn't blacklist Anthropic.

Tuesday, March 3: Altman's all-hands and public protest outside OpenAI HQ

Roughly 40 to 50 people gathered to air their grievances, which extended beyond the deal with the Pentagon.

Manuel Orbegozo for BI

Hours after Altman held an all-hands meeting with OpenAI employees to discuss the agreement, protesters arrived at the AI company's headquarters.

Roughly 40 to 50 people gathered to air their grievances, which extended beyond the deal with the Pentagon.

Read the original article on Business Insider

Read full story at source