SUBSCRIBE TO OUR FREE NEWSLETTER
Daily news & progressive opinion—funded by the people, not the corporations—delivered straight to your inbox.
5
#000000
#FFFFFF
5
#000000
#FFFFFF
openai
Watchdog Says Trump-Backed AI Data Center Push 'Raises Massive Antitrust Concerns'
"Even if it turns out to be structured to avoid antitrust law enforcement, it plainly will concentrate power in a small number of corporate hands," said Public Citizen co-president Robert Weissman.
Jan 22, 2025
U.S. President Donald Trump on Tuesday highlighted a new private-sector initiative to invest as much as $500 billion over four years into developing infrastructure to support artificial intelligence, starting with a raft of power-intensive data centers in Texas. The move drew swift criticism from one watchdog group on antitrust and environmental grounds.
The initiative, Stargate, is a joint venture of the tech firms OpenAI, Oracle, and SoftBank. Trump hosted the leaders of those companies—OpenAI CEO Sam Altman, Oracle Chairman Larry Ellison, and SoftBank CEO Masayoshi Son—at the White House to announce the initiative just one day after he signed an executive order rolling back a Biden-era executive order implemented in 2023 that sought to put safeguards on AI.
"I think this will be the most important project of this era," said Altman, according to the Washington Post. "We wouldn't be able to do this without you, Mr. President," he added, though both the Post and and the Associated Press noted that the creation of the partnership predated Trump's return to the White House.
Biden's 2023 executive order on AI placed safety obligations on AI developers and called on federal agencies to examine the technology's risks. But Biden, too, was interested in boosting AI infrastructure development. Right before he departed, in early mid-January, Biden signed an executive order directing federal agencies to identify government sites that could be leased to private companies for the construction of AI data centers.
Environmental groups and tech advocacy groups have long advocated for greater safeguards on AI, pointing to the technology's potential impact on the climate emergency.
The average query in the AI-powered chatbot ChatGPT requires 10 times the amount of energy a Google search needs, and "in that difference lies a coming sea change in how the U.S., Europe, and the world at large will consume power—and how much that will cost," according to a 2024 analysis published by the investment firm Goldman Sachs. Goldman Sachs analysts believe that AI will represent about 19% of data center power demand by 2028.
AI infrastructure is also water intensive. Global AI demand is projected to require more water extraction in a year than the country of Denmark by 2027, according to one study.
"The alarming surge in these centers' energy demand is on track to extend the fossil fuel era... [and] it is already increasing costs for some consumers and threatens to bring about a larger affordability crisis, while lining the pockets of Big Tech billionaires," said Karen Orenstein, a director at the environmental group Friends of the Earth, following Biden's January executive order. "For the sake of our planet and its people, we need to rein in Big Tech and regulate AI," she said.
Meanwhile, the joint venture to build out AI infrastructure has also drawn scrutiny from one watchdog group over concerns of corporate concentration.
Public Citizen co-president Robert Weissman said Wednesday that "the new Stargate plan—at minimum—raises massive antitrust concerns. Even if it turns out to be structured to avoid antitrust law enforcement, it plainly will concentrate power in a small number of corporate hands."
"Absent a commitment to bring on new, renewable energy to power an even greater spike in AI power demand, the Stargate build out threatens to worsen the rush to climate catastrophe and to drive up consumer electric bills," he added.
Another observer, Jeffrey Westling of the American Action Forum, remarked on the timing of the announcement.
"Interesting to wait to announce this until the Trump Admin. Assuming its all private investment, maybe they were worried about FTC/DOJ antitrust scrutiny?" he wrote on X Tuesday.
Keep ReadingShow Less
'Hellscape': Microsoft Deal Would Reopen Three Mile Island Nuclear Plant to Power AI
"This is another chapter in a nightmare that won't end," a campaigner said.
Sep 20, 2024
The corporation that owns the shuttered nuclear plant on Three Mile Island on Friday announced a deal with Microsoft to reopen the facility to provide power to the tech company for data centers using artificial intelligence.
Three Mile Island is well-known as the site of largest nuclear disaster in U.S. history—a reactor there, Unit 2, partially melted down in 1979. However, the site's other reactor, Unit 1, continued to operate safely until 2019, when it was closed for economic reasons.
With the help of tax breaks from the Inflation Reduction Act (IRA), plant owner Constellation Energy plans to spend $1.6 billion to restart Unit 1, with all of the power going to Microsoft for the first 20 years. Microsoft and other tech firms use inordinate amounts of energy to power data centers used for AI and have advocated for nuclear as a zero-emissions power source.
Though it doesn't emit carbon, nuclear power's downsides make it the subject of fierce opposition from many environmental and public safety groups.
Friday's deal, combining nuclear power and AI, which also poses great safety risks, was too much for The Philadelphia Inquirer columnist Will Bunch, who quipped that "the hellscape of modern life" had been captured in one headline.
Local campaigners vowed to push against the reopening and keep the area free of nuclear activity.
"We will challenge this proposal at every venue that is available for us," Eric Epstein, a former chairman of Three Mile Island Alert, a campaign group, told the Inquirer.
"This is another chapter in a nightmare that won't end," he added.
Siri, define the hellscape of modern life in one headline https://t.co/miQMSWROmp
— Will Bunch (@Will_Bunch) September 20, 2024
Three Mile Island would be the first decommissioned U.S. nuclear plant to reopen, and the first to provide all of its power to one corporation, according toThe New York Times. Microsoft and Constellation didn't disclose the financial details of the deal.
About 19% of electricity in the U.S. comes from nuclear power, and a drive for "clean energy," as well as the IRA credits, have spurred growth in the sector. Microsoft co-founder and former CEO Bill Gates is a vocal proponent and has started his own nuclear company, TerraPower, which is building a plant in Wyoming.
The Three Mile Island project, expected to get the plant back online in 2028, still needs regulatory approval at multiple levels. Pennsylvania Gov. Josh Shapiro, a Democrat, has already come out in support of the plan.
A study commissioned by the Pennsylvania Building & Construction Trades Council, which represents more than 115 local unions, found that reopening the plant would create 3,400 direct and indirect jobs, including 600 at the plant itself, which is in the middle of the Susquehanna River, just south of Harrisburg, the state capital. The plant's reopening is seen by some community leaders as the revival of an "economic anchor in a region beset with financial hardship," according toThe Washington Post.
Pennsylvania has five active nuclear plants, including two owned by Constellation, whose stock price shot up on Friday morning after the Three Mile Island announcement was made. The company and other industry backers celebrated the symbolic victory of restarting the plant.
"If anything says nuclear power is here to stay and expand, it's Three Mile Island reopening!" Amir Adnani, CEO of Uranium Energy Corp, wrote on social media.
Epstein, the campaigner, said the focus should be finishing the cleanup from the 1979 disaster. About 99% of the Unit 2's fuel has been removed to Idaho, but the last 1% has proven difficult to deal with.
"First things first, remove the waste from the island, and clean up [Unit 2]," Epstein said.
Keep ReadingShow Less
Israeli Government Funded Covert Influence Campaign Targeting US Lawmakers: NYT
The head of an Israeli watchdog group called the operation "anti-democratic" and "extremely irresponsible."
Jun 05, 2024
Israel's Ministry of Diaspora Affairs organized and paid for a digital campaign to influence U.S. lawmakers, especially Democrats who are Black, The New York Timesreported on Wednesday.
The ministry allotted $2 million to the operation in October and hired Stoic, a Tel Aviv-based political marketing firm, to carry it out. Stoic established fake news websites and hundreds of fake accounts on X, Instagram, and Facebook that posted pro-Israeli messages, trying to push lawmakers such as Rep. Hakeem Jeffries (D-N.Y.), the House minority leader, Rep. Ritchie Torres (D-N.Y.), and Sen. Raphael Warnock (D-Ga.) to fund Israel's military and support its war efforts, the Times reported.
The influence campaign had been reported by a few news and nonprofit organizations in recent months, but the Times article, which drew from operation documents and interviews with current and former diaspora ministry officials, was the first to show that Israel's government was behind it. Haaretz, an Israeli newspaper, published a related story one hour later on Wednesday.
Critics condemned the Israeli government for its role in the disinformation campaign.
"So in addition to the pro-Israel lobby spending tens of millions to defame and defeat progressives in Congress, we now learn that Israel creates fake media to target friends and opponents by inundating with fake news supporting Israeli positions," James Zogby, co-founder of the Arab American Institute, wrote on social media.
Wow—New York Times & Haaretz report Israel ordered a secret digital operation on US lawmakers (esp. Black Democrats) that spread misinformation & anti-Arab sentiment and attacked pro-Palestinian Americans, to sway public opinion & influence the lawmakers to fund Israel’s military pic.twitter.com/D5U5ewiqLP
— Prem Thakker (@prem_thakker) June 5, 2024
The disinformation campaign comes amid other efforts by pro-Israel groups to influence U.S. politics during its assault on Gaza, notably the lobbying and campaign money spent by groups such as the American Israel Public Affairs Committee (AIPAC) and its affiliates.
The Israeli disinformation campaign also drew comparisons to Russia's well-known attempt to interfere in the 2016 U.S. presidential election, which was a central focus of the U.S. political commentariat in the years that followed. Ishmael Daro, an editor at Democracy Now!, made a tongue-in-cheek prediction that the reaction from the U.S. political establishment would be similar this time.
When Russia set up bots to post ineffective propaganda in 2016, it led to a multi-year meltdown by much of the U.S. political establishment that basically treated it like a coup attempt. I assume we'll get a similar reaction to this. https://t.co/sEXeO41ozb pic.twitter.com/nPkQtWsCyb
— ishmael n. daro (@iD4RO) June 5, 2024
Last week, both Meta and OpenAI issued reports on Stoic's disinformation campaign and said they had blocked the company's network from further activity. Meta said it had closed more than 500 fake Facebook accounts and OpenAI called Stoic a "for-hire Israeli threat actor," NBC Newsreported. Stoic's users remain active on X, the Times reported.
Many of the fake social media posts were generated using ChatGPT, the AI-powered chatbot owned by OpenAI, and much of the language in the posts was "stilted" and repetitive, the Times reported.
The covert scheme has also been characterized as "sloppy" and "ineffective," and it made little penetration with the general public or government figures. "We found and removed this network early in its audience building efforts, before they were able to gain engagement among authentic communities," Meta wrote in its report.
The Times did not explain that the covert influence campaign was discovered in February by the Digital Forensic Research Lab (DFRLab) of the Atlantic Council, a Washington, D.C.-based think tank, and by Marc Owen Jones, a professor in Middle East studies and digital humanities at Hamad Bin Khalifa University in Qatar, according to social media posts.
FakeReporter, an Israeli disinformation watchdog, followed up those initial discoveries with a March report on the campaign's activities, including the fake social media accounts and creation of the online platforms—Non-Agenda, The Moral Alliance, and Unfold Magazine—that created or republished news from a pro-Israel perspective, focusing on, for example, purported links between the United Nations Relief and Works Agency (UNRWA) and Hamas. The findings were reported in Haaretz at the time.
That Israel "ran an operation that interferes in U.S. politics is extremely irresponsible," Achiya Schatz, the executive director of FakeReporter, told the Times. He characterized it to Haaretz as "amateurish" and "anti-democratic."
FakeReporter in fact issued a second report on Wednesday showing that Stoic's influence network may have gone further than the Times reporting shows. The watchdog group uncovered four additional websites, apparently Stoic-affiliated, that contain Islamophobic and anti-immigrant content. DFRLab had issued a report in March which also cited pro-Israeli disinformation and Islamophobic rhetoric, in that case targeted largely at Canadians.
The new report concluded that the influence network has "apparently developed into a large-scale effort to target various groups, some outside the U.S., using Islamophobic and anti-immigrant content."
Keep ReadingShow Less
Most Popular