Connect with us

News

Where have OpenAI’s founders gone?

Published

on

Where have OpenAI’s founders gone?

Just two of OpenAI’s 11-strong founding team are still active at the ChatGPT maker, after an exodus following November’s attempted boardroom coup against chief executive Sam Altman.

Three co-founders have departed so far this year, including John Schulman, who defected to its artificial intelligence rival Anthropic this week. Greg Brockman, OpenAI’s president, also said on Monday he would be taking extended leave from the company.

A high rate of turnover is not unusual at a start-up. However, attrition of senior figures at OpenAI has stepped up in recent months following November’s leadership crisis, when Altman was fired by his board only to be reinstated days later.

Since then, the loss of executives and staffers working on AI safety and research has raised questions about the direction of the $86bn company, which is in a fierce battle to stay ahead of rivals including Google and Anthropic.

Some co-founders have absconded to rivals, others to launch their own AI companies, while the team’s most famous former member — Elon Musk — has become a vociferous critic of OpenAI in public and in the courts.

Advertisement

OpenAI had a larger number of founders than most Silicon Valley start-ups because Altman and Brockman wanted to build an AI supergroup of the top researchers in the field when it got started in 2015. Here is where those 11 founders are now.

Leavers

Greg Brockman

on a leave of absence since August 2024

Brockman is a core member of OpenAI’s founding team. He was persuaded by Altman and Musk to leave his job as chief technology officer at financial technology company Stripe and take on the same position at OpenAI.

He has been a key Altman ally since the beginning. When the board moved against Altman in a coup in November, Brockman was also removed as a director. The two returned to their posts together when the board backtracked five days later.

On Monday, the company’s president announced he would be taking a sabbatical for the rest of the year.

“First time to relax since co-founding OpenAI 9 years ago,” he wrote on X. “I’ve poured my life for the past 9 years into OpenAI, including the entirety of my marriage. Our work is important to me, but so is life.”

Advertisement

John Schulman

joined Anthropic in August 2024

Schulman, a research scientist who played a vital role in building the company’s ChatGPT chatbot, announced he would leave OpenAI on Monday. He was responsible for fine-tuning the company’s AI models and ensuring they behaved in a way that conformed to human values — a process known as alignment.

He will take up a similar role at rival start-up Anthropic, which itself was founded by ex-OpenAI researchers in 2021.

“This choice stems from my desire to deepen my focus on AI alignment, and to start a new chapter of my career where I can return to hands-on technical work, alongside people deeply engaged with the topics I’m most interested in,” Schulman said in a note to colleagues on Monday.

Ilya Sutskever

left to found Safe Superintelligence in May 2024

Sutskever left his position as OpenAI’s chief scientist six months after voting with the company’s board to remove Altman. Sutskever, one of the most prominent researchers in the field, reversed his position and backed the chief executive’s return a few days later.

Nonetheless, he has been largely absent from public view in the months since the abortive coup, and in May he left to start a company called Safe Superintelligence.

Andrej Karpathy

left to found Eureka Labs in February 2024

Karpathy, a research scientist who was advised at Stanford by “Godmother of AI” Fei-Fei Li, first left OpenAI in 2017 to join Tesla as a senior director. He returned to OpenAI in 2023 and left again a year later to launch Eureka Labs, which is building AI teaching assistants.‎

Durk Kingma

left for Google Brain in June 2018

Kingma, who worked on developing algorithms for generative AI models, left for Google in the summer of 2018. He has continued to lead research on large language models and image models at Google Brain, which merged with DeepMind last year.

Elon Musk

resigned from the board in 2018

Musk, who provided much of OpenAI’s early funding, left the company in 2018 after clashing with Altman over the direction of research. The billionaire launched a rival company, xAI, last year and claims he can overhaul OpenAI’s lead.

The Tesla, SpaceX and X chief has also launched a number of lawsuits against Altman and OpenAI, arguing this week that he was induced to invest in the AI company by its “fake humanitarian mission”.

Pamela Vagata

JOINED STRIPE in 2016

Vagata, listed as a founding member of OpenAI in the company’s launch announcement, makes no mention of the start-up in her LinkedIn profile. She joined Stripe as a technical leader in the fintech company’s AI team in 2016, and founded early-stage venture capital firm Pebblebed in 2021.

‎Vicki Cheung

JOINED LYFT in 2017

Cheung, who worked on language-learning app Duolingo before becoming OpenAI’s first engineer, left the company in 2017 to join ride-hailing start-up Lyft. In 2020 she founded machine learning start-up Gantry alongside former OpenAI researcher Josh Tobin.

Trevor Blackwell

left in 2017

Blackwell was a partner at Y Combinator, the San Francisco start-up accelerator that Altman ran before establishing OpenAI. He helped launch the AI company and left in 2017. A robotics enthusiast, he is now based in Gloucestershire, England.

‎Remainers

Sam Altman

Altman remains as OpenAI’s chief executive after surviving a boardroom coup in November, during which directors accused him of not being “consistently candid” with them. He was reinstated five days after being fired on the back of a campaign by employees and investors in OpenAI, including Microsoft.

The departure of other senior figures has left the 39-year-old as by far the most prominent figure at the company, and the reconstitution of the board after its failed ousting has further solidified his power.

Advertisement

Wojciech Zaremba

Polish computer scientist Zaremba remains at OpenAI where he works as a researcher. He called on the board to resign after they moved against Altman, and has since urged his chief executive and Musk to drop their “unnecessary fight”.

“It would be so much better to put your creative energy into building the future you dream of over a quarrel. May you (both) be happy and find peace,” he wrote in a post on X in March, signing off with a love heart.

Additional reporting by Madhumita Murgia

News

Anthropic CEO says he’s sticking to AI “red lines” despite clash with Pentagon

Published

on

Anthropic CEO says he’s sticking to AI “red lines” despite clash with Pentagon

Hours after a bitter feud between the Pentagon and Anthropic ended with the Trump administration cutting off the artificial intelligence startup, Anthropic CEO Dario Amodei told CBS News in an exclusive interview Friday night he wants to work with the military — but only if it addresses the firm’s concerns.

“We are still interested in working with them as long as it is in line with our red lines,” he said.

The conflict centers on Anthropic’s push for guardrails that explicitly prevent the military from using its powerful Claude AI model to conduct mass surveillance on Americans or to power autonomous weapons. The Pentagon wants the ability to use Claude for “all lawful purposes,” and says it isn’t interested in either of the uses that Anthropic was concerned about.

The military gave Anthropic a Friday evening deadline to either meet its demands or get cut off from its lucrative Defense Department contracts. With the two sides still seemingly still far apart, President Trump on Friday ordered federal agencies to “immediately” stop using Anthropic’s technology. Then, Defense Secretary Pete Hegseth declared the company a “supply chain risk,” directing military contractors to also stop working with the AI startup.

In his interview later Friday, Amodei stood by the guardrails sought by Anthropic, which is the only company whose AI model is deployed on the Pentagon’s classified networks.

Advertisement

“Our position is clear. We have these two red lines. We’ve had them from day one. We are still advocating for those red lines. We’re not going to move on those red lines,” Amodei later said. “If we can get to the point with the department where we can see things the same way, then perhaps there could be an agreement. For our part and for the sake of U.S. national security, we continue to want to make this work.”

Amodei told CBS News that Anthropic has sought to deploy its AI models for military use because “we are patriotic Americans” and “we believe in this country.” But the company is worried that some potential uses of AI could clash with American values, he said.

Mass surveillance is a risk, Amodei argued, because “things may become possible with AI that weren’t possible before,” and the technology’s potential is “getting ahead of the law.” He warned that the government could buy data from private firms and use AI to analyze it.

In theory, artificial intelligence could also be used to power fully autonomous weapons that select targets and carry out strikes without any human input. Amodei said his company isn’t categorically opposed to those kinds of weapons, especially if U.S. adversaries develop them, but “the reliability is not there yet” and “we need to have a conversation about oversight.”


The Free Press: Will AI Doom Us All? The Market Can’t Decide

Advertisement

Since AI technology is still unpredictable, Amodei is concerned that autonomous weapons could target the wrong people by mistake. And unlike with human-powered weaponry, it’s not clear who is responsible for the decisions made by fully autonomous weapons.

“We don’t want to sell something that we don’t think is reliable, and we don’t want to sell something that could get our own people killed or that could get innocent people killed,” he said.

Amodei called the guardrails around surveillance and autonomous weapons “narrow exceptions,” and said the company has no evidence that the military has run into either of them.

The Pentagon’s position is that federal law already prevents it from surveilling Americans en masse, and fully autonomous weapons are already restricted by internal military policies, so there is no need to put restrictions on those uses of AI in writing.

Emil Michael, the Pentagon’s chief technology officer, told CBS News in an interview Thursday: “At some level, you have to trust your military to do the right thing.”

Advertisement

“But we do have to be prepared for the future. We do have to be prepared for what China is doing,” Michael said, referring to how U.S. adversaries use AI. “So we’ll never say that we’re not going to be able to defend ourselves in writing to a company.” 

As a compromise, Michael said the military had offered written acknowledgements of the federal laws and military policies that restrict mass surveillance and autonomous weapons — though Anthropic said that offer was “paired with legalese” that allowed the guardrails to be ignored.

As the conflict between Anthropic and the Pentagon escalated this week, top military officials accused the company and Amodei of trying to impose their values onto the government. Hegseth called Anthropic “sanctimonious” and arrogant, Michael said that Amodei has a “God-complex” and Mr. Trump called the AI startup a “radical left, woke company.”

“Their true objective is unmistakable: to seize veto power over the operational decisions of the United States military. That is unacceptable,” Hegseth alleged.

Said Mr. Trump: “Their selfishness is putting AMERICAN LIVES at risk, our Troops in danger, and our National Security in JEOPARDY.”

Advertisement

Asked if weighty questions about AI guardrails should be left up to Anthropic rather than the government, Amodei told CBS News that “one of the things about a free market and free enterprise is, different folks can provide different products under different principles.”

He also said: “I think we are a good judge of what our models can do reliably and what they cannot do reliably.”

In the long run, he said, Congress should probably weigh in on AI safeguards.

“But Congress is not the fastest moving body in the world. And for right now, we are the ones who see this technology on the front line,” said Amodei.

With Anthropic and the Pentagon unable to reach a deal by Friday, the military is now expected to phase out its use of Anthropic’s AI technology within six months and transition to what Hegseth called “a better and more patriotic service.”

Advertisement

Hegseth also labeled Anthropic a “supply chain risk” and said all companies that do business with the military are now expected to cut off “any commercial activity with Anthropic.” 

Amodei called that an “unprecedented” move for an American firm rather than a foreign adversary, and he said the government’s statements have been “retaliatory and punitive.” And he argued that Hegseth doesn’t have the legal authority to bar all military contractors from working with Anthropic, and can only stop them from using Anthropic for government contracts.

He also said that Anthropic hasn’t formally received any information from the Pentagon informing it of a supply chain risk designation, but “when we receive some kind of formal action, we will look at it, we will understand it and we will challenge it in court.”

Asked if he has a message for the president, Amodei said “everything we have done has been for the sake of this country” and “for the sake of supporting U.S. national security.”

“Disagreeing with the government is the most American thing in the world,” he said. “And we are patriots. In everything we have done here, we have stood up for the values of this country.”

Advertisement

Continue Reading

News

How the federal government is painting immigrants as criminals on social media

Published

on

How the federal government is painting immigrants as criminals on social media

Getty Images, Dept. of Homeland Security and The White House via X/Collage by Emily Bogle/NPR

Two days after At Chandee, who goes by Ricky, was arrested by U.S. Immigration and Customs Enforcement, the White House’s X account posted about him, calling the 52-year-old the “WORST OF WORST” and a “CRIMINAL ILLEGAL ALIEN.”

Except that the photo the White House posted was of a different person. The post also incorrectly claimed Chandee had multiple felony convictions — he has one, for second-degree assault in 1993 when he was 18 years old. He shot two people in the legs and served three years in prison.

Advertisement
At "Ricky" Chandee with his wife, Tina Huynh-Chandee.

At “Ricky” Chandee with his wife, Tina Huynh-Chandee.

Via the Chandee family


hide caption

toggle caption

Via the Chandee family

Advertisement

Chandee, who came to the U.S. as a child refugee, was ordered to be deported back to his home country, Laos. But Laos had not been accepting all of the people the U.S. wanted it to, so the federal government determined that it was likely infeasible to deport him, his lawyer Linus Chan told NPR. Chandee therefore was granted permission to stay in the U.S. and work so long as he checked in with immigration authorities periodically. He has not missed a check-in in over 30 years and has not had another criminal incident.

People who know Chandee do not see him as “worst of the worst.”

After Chandee completed his prison sentence, he finished school and became an engineering technician. He worked for the City of Minneapolis for 26 years, became a father, and his son grew up to join the military.

In his free time, Chandee enjoys hiking and foraging for mushrooms, Minnesota Public Radio reported.

Advertisement

“We are proud to work alongside At ‘Ricky’ Chandee,” said Tim Sexton, Director of Public Works for the City of Minneapolis in a statement. “I don’t understand why he would be a target for removal now, why he was brutally detained and swiftly flown to Texas, or how his removal benefits our city or country.” Chandee is petitioning for his release in federal court.

Chandee’s case is not unique 

Social media accounts from the White House, the Department of Homeland Security and other immigration agencies have spent much of the past year posting about people detained in the administration’s immigration crackdown, typically portraying them as hardened, violent criminals. That’s even as over 70% of the people detained don’t have criminal records according to ICE data.

NPR’s research of cases in Minnesota shows that while many of the people who have been highlighted on social media do have recent, serious criminal records, about a quarter are like Chandee, with decades-old convictions, minor offenses or only pending criminal proceedings. Scholars of immigration, media and criminal law say such a media campaign is unprecedented and paints a distorted picture of immigrants and crime.

A year into President Trump’s second term, the X accounts of DHS and ICE have posted about more than 2,000 people who were targets of mass deportation efforts. Starting late last March, DHS and ICE began posting on X on a near daily basis, often highlighting apprehensions of multiple people a day, an NPR review of government social media posts show.

Advertisement

Among the 2,000 people highlighted by the agencies, NPR identified 130 who were arrested by federal agents in Minnesota and tried to verify the government’s statements about their criminal histories.

In most of the social media posts, the government did not provide the state where the conviction occurred or the person’s age. Public court records do not tend to include photos so definitive identification can be a challenge.

NPR derived its findings from cases where it was able to locate a name and matching criminal history in the Minnesota court and detention system, in nationwide criminal history databases, sex offender databases, and in some cases, federal courts and other state courts.

In 19 of the 130 cases, roughly 1-in-7, public records show the most recent convictions were at least 20 years ago.

Seventeen of the 19 cases with old convictions did include violent crimes like homicide and first-degree sexual assault. ICE provided some of those names to Fox News as key examples of the agency’s accomplishments. “It’s the most disturbing list I’ve ever seen,” said Fox News reporter Bill Melugin on X, highlighting the criminal convictions of each person on the list.

Advertisement

For seven people, their only criminal history involved driving under the influence or disorderly conduct.

ICE agents approach a house before detaining two people in Minneapolis on Jan. 13.

ICE agents approach a house before detaining two people in Minneapolis on Jan. 13.

Stephen Maturen/Getty Images


hide caption

toggle caption

Advertisement

Stephen Maturen/Getty Images

Six of the 130 Minnesota cases highlighted by the administration involved people with no criminal convictions. The government’s social media posts for those six instead rely upon the charges and arrests as evidence of their criminality, even though arrests don’t always lead to charges and charges can be dismissed.

In yet another case, the government highlighted a criminal charge even while noting it had been dismissed. (The person did have other existing convictions.)

For 37 of the 130 people, NPR was unable to confirm matching criminal history after consulting the databases and news coverage. Some of the names turned up no criminal history at all. The government said these people committed crimes ranging from homicide and assault to drug trafficking, and cited one by name to Fox News. NPR tried to reach out to all 37 people and their families for comment but did not receive a response from any.

Advertisement

In a statement to NPR, DHS’s chief spokesperson Lauren Bis did not dispute NPR’s findings or provide documentation where NPR wasn’t able to confirm matching criminal history.

“The fact that NPR is defending murderers and pedophiles is gross,” Bis wrote. “We hear far too much about criminals and not enough about their victims.” before listing four of the people with old convictions of homicide and sexual assault, underlining the date of deportation order for three of them.

Images designed to trigger emotion

The stream of social media posts with photos of mostly nonwhite people are meant to draw an emotional response, says Leo Chavez, an emeritus professor of anthropology at the University of California, Irvine. They “have been used repeatedly over and over to get people to buy into, really drastic, drastic and draconian actions and policies,” he said.

Chavez, whose most recent book is The Latino Threat: How Alarmist Rhetoric Misrepresents Immigrants, Citizens, and the Nation, recalls how political campaigns in past decades presented images of Latinos — often men — without context. “Just by showing their image, showing brown people, particularly brown men, it’s supposed to be scary.”

The fact that the government’s social media posts come with statements about criminal history as well as photos reinforces that emotional response, Chavez said. DHS has previously acknowledged inaccuracies on their website. But even if the department issues corrections, Chavez said, “the goal was actually achieved, which was to reinforce the criminality and the visualization.”

Advertisement

CNN’s analysis of DHS’s “Arrested: Worst of the Worst” website showed that for hundreds out of about 25,000 people posted on the website, the crimes listed were not violent felonies. Instead, DHS listed people with records that included traffic offenses, marijuana possession or illegal reentry. DHS said the website had a “glitch” that it will fix but also that the people in question “have [committed] additional crimes.”

“I’ve never seen anything like this when it comes to immigration enforcement in the modern era,” said Juliet Stumpf, a professor at Lewis & Clark Law School who studies the intersection of immigration and criminal law. She said the drumbeat of social media posts focused on specific individuals was like “FBI’s most wanted posters” or “like reality TV shows.”

Then-DHS Assistant Secretary for Public Affairs Tricia McLaughlin, flanked by deputy director of U.S. Immigration and Customs Enforcement Madison Sheahan, left, and Acting director of U.S. Immigration and Customs Enforcement Todd Lyons, speaks during a news conference at ICE Headquarters, in Washington, D.C., on May 21, 2025.

Then-DHS Assistant Secretary for Public Affairs Tricia McLaughlin, flanked by deputy director of U.S. Immigration and Customs Enforcement Madison Sheahan (left), and Acting director of U.S. Immigration and Customs Enforcement Todd Lyons, speaks during a news conference at ICE Headquarters, in Washington, D.C., on May 21, 2025.

Jose Luis Magana/AP


hide caption

Advertisement

toggle caption

Jose Luis Magana/AP

Stumpf drew a parallel with an incident from the 1950s when the U.S. government deported two permanent residents suspected of being communists. “The government was kind of proclaiming and celebrating their deportation because getting rid of these communists was making the country safer,” said Stumpf, “Maybe that’s comparable to something like [this].”

An analysis by the Deportation Data Project shows a dramatic increase in arrests of noncitizens without criminal records during President Trump’s current term compared to President Biden’s term.

Advertisement

“If you look at research, immigrants actually tend to commit fewer crimes than even U.S. citizens do. And that’s true of immigrants who have lawful status here and immigrants who don’t,” said Stumpf. “If we have a number of social media posts that are painting immigrants as the worst of the worst…it’s actually really putting out a distorted version of reality about who immigrants actually are.”

Some claims are disputed by other authorities

In some posts, DHS and ICE have also used photos of people and statements about their criminal histories to burnish the federal government’s accomplishments, defend their agents and criticize states like Minnesota. State and local authorities have in turn pushed back, and some of the federal government’s claims about the people it has detained have been met with setbacks in the courts.

DHS accused Minnesota’s Cottonwood County of not honoring detainers, written requests by ICE to hold prisoners in custody for a period of time so ICE can pick them up. In one post, the agency identified a person who was charged with child sexual abuse, writing “This is who sanctuary city politicians and anti-ICE agitators are defending.”

The Cottonwood County sheriff’s office said DHS’s post “misrepresented the truth” in their own post on Facebook. According to their account, the county did honor the detainer but ICE said it was unable to pick up the person before the order expired and the county had to release the suspect.

The Minnesota Department of Corrections wrote in a blog post that dozens of people DHS listed on its “Worst of the Worst” website were not arrested as DHS described, but were transferred to ICE by the state because they were already in state custody. The Corrections Department has since launched a page dedicated to “correct the Department of Homeland Security’s (DHS) repeated false claims.”

Advertisement

The “Worst of the Worst” website has some overlap with the department’s social media posts, but it contains a much larger number of people — over 30,000 nationally. It included a Colombian soccer star who was extradited to the U.S., tried in Texas, convicted of drug trafficking and served time in federal prison. The website incorrectly describes him as being arrested in Wisconsin. The soccer player, Jhon Viáfara Mina, recently finished his sentence early and returned to Colombia, according to Spanish newspaper El Diario Vasco.

In some instances, DHS and ICE wrote about incidents where they ran into conflict when carrying out arrests. In those posts, they named the arrestees and posted their photos. But in one case where the incident went to court, the government’s account of the events shifted. After a federal agent shot Julio C. Sosa-Celis in Minneapolis in January, DHS claimed he was lodging a “violent attack on law enforcement.” Assault charges against Sosa-Celis fell apart in court as new evidence surfaced, and the officers involved were put on leave.

Despite the fact that the charges were dropped, DHS’s post profiling Sosa-Celis remains online.

Continue Reading

News

Bill Clinton to testify before House committee investigating Epstein links

Published

on

Bill Clinton to testify before House committee investigating Epstein links

Former president Bill Clinton is scheduled to give deposition Friday to a congressional committee investigating his links to Jeffrey Epstein, one day after Hillary Clinton testified before the committee and called the proceedings “partisan political theatre” and “an insult to the American people”.

During remarks before the House oversight committee, Hillary Clinton, the former secretary of state, insisted on Thursday that she had never met Epstein.

The former Democratic president, however, flew on Epstein’s private jet several times in the early 2000s but said he never visited his island.

Clinton, who engaged in an extramarital affair while president and has been accused of sexual misconduct by three women, also appears in a photo from the recently released files, in a hot tub with Epstein and a woman whose identity is redacted.

Clinton has denied the sexual misconduct claims and was not charged with any crimes. He also has not been accused of any wrongdoing connected to Epstein.

Advertisement

Epstein visited the White House at least 17 times during the early years of Clinton’s presidency, according to White House visitor records cited in news reports. Clinton said he cut ties with him around 2005, before the disgraced financier, who died from suicide in 2019, pleaded guilty to solicitation of a minor in Florida.

The House committee subpoenaed the Clintons in August. They initially refused to testify but agreed after Republicans threatened to hold them in contempt.

The Clintons asked for their depositions to be held publicly, with the former president stating that to do so behind closed doors would amount to a “kangaroo court”.

“Let’s stop the games + do this the right way: in a public hearing,” Clinton said on X earlier this month.

The committee’s chair, James Comer, did not grant their request, and the proceedings will be conducted behind closed doors with video to be released later.

Advertisement

On Thursday, Hillary Clinton’s proceedings were briefly halted after representative Lauren Boebert leaked an image of Clinton testifying.

During the full day deposition, Clinton said she had no information about Epstein and did not recall ever meeting him.

Before the deposition, Comer said it would be a long interview and that one with Bill Clinton would be “even longer”.

Continue Reading

Trending