xAI announces $200m US military deal after Grok chatbot had Nazi meltdown
The DoD on Monday also announced similar contracts with $200m ceilings with several other major US-based artificial intelligence developers, including Google, Anthropic and OpenAI. The agency is partnering with the General Services Administration to make these companies' AI tools available for use throughout the federal government.
'Leveraging commercially available solutions into an integrated capabilities approach will accelerate the use of advanced AI as part of our joint mission-essential tasks in our warfighting domain as well as intelligence, business, and enterprise information systems,' the US chief digital and AI officer Dr Doug Matty said in a statement.
The contracts deepen the US military's ties with AI developers and are poised to expand the use of artificial intelligence within the US government after Musk's so-called 'department of government efficiency' (Doge) oversaw mass firings of workers throughout federal agencies. Until Musk's recent falling out with Donald Trump, the xAI founder was de facto leader of Doge as it gutted government agencies and in some cases pushed for departments to use the Grok chatbot.
xAI's contract announcement comes after the company was forced to issue a public apology after Grok posted a string of responses on X last week that included promoting Nazi ideology and rape fantasies. The company claimed that it fixed the issue and subsequently unveiled its latest AI model with a $300 a month subscription for an advanced version of the tool.
The DoD's contract will give xAI a boost of revenue as it seeks to compete with more established AI developers like OpenAI, which is led by Musk's former associate turned rival, Sam Altman. Musk has been heavily promoting xAI and attempting to use other parts of his tech empire to support its future, including having SpaceX invest $2bn into the startup, allowing it to acquire X, formerly, Twitter, and announcing on Sunday that Tesla shareholders will vote on their own investment in xAI.
xAI announced the deal and the creation of what it calls 'Grok For Government' in a post on its website on Monday, detailing that in addition to its publicly available products it would create custom AI-powered applications for potential use in healthcare, national security and other public services.
'Under the umbrella of Grok For Government, we will be bringing all of our world-class AI tools to federal, local, state, and national security customers,' xAI said in a statement on its website. 'These customers will be able to use the Grok family of products to accelerate America – from making everyday government services faster and more efficient to using AI to address unsolved problems in fundamental science and technology.'
Sign up to TechScape
A weekly dive in to how technology is shaping our lives
after newsletter promotion
Musk has long complained that AI chatbots are designed to promote 'woke' ideology and vowed his Grok product would be 'maximally truth seeking'. It has repeatedly run into controversy over promoting conspiracies and falsehoods, including earlier this year giving unprompted responses that made false claims of 'white genocide' taking place in South Africa – echoing claims that Musk has made himself.
Ethics watchdogs, Democratic lawmakers and privacy advocates have expressed concerns about how Musk and Doge have implemented AI in government and gained access to sensitive data while embedded at government agencies. Doge staffers previously fed government data into a custom version of Grok's chatbot in a potential violation of privacy and security laws, Reuters reported in May.
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Reuters
42 minutes ago
- Reuters
BlackRock's assets hit record $12.53 trillion on second-quarter market rally
July 15 (Reuters) - BlackRock's (BLK.N), opens new tab assets under management hit a new high in the second quarter as global markets rallied on the prospect of trade deals and interest-rate cuts from the U.S. Federal Reserve, brushing aside earlier tariff-related jitters. Shares of the company rose 1.1% in premarket trading. A robust labor market, a healthy consumer and hopes that President Donald Trump would ease some of his harsher trade measures pushed major U.S. indices to all-time highs through the end of June. That marked a sharp reversal from early April, when tumult in U.S. trade and geopolitical policy battered confidence and fueled recession fears, concerns that BlackRock CEO Larry Fink echoed at the time. The benchmark S&P 500 (.SPX), opens new tab index rose 10.57% in the second quarter of 2025 after escaping bear market territory. BlackRock's assets under management rose to $12.53 trillion in the quarter ended June 30, from $10.65 trillion last year. However, long-term net inflows fell to $46 billion in the quarter, down 9.8%. "We generated 6% organic base fee growth for the second quarter and the first half of 2025 and 7% over the last 12 months," Fink said. As equities rallied, fixed-income products saw outflows of $4.66 billion, BlackRock said. Trends for the business are being closely scrutinized, given the turbulence in U.S. treasuries this quarter. The benchmark 10-year yield recorded one of its biggest weekly increases since 2001 after the "Liberation Day" shock. BlackRock's fixed-income executives expressed concerns last month that ballooning U.S. debt could suppress appetite for longer-dated treasuries and the dollar, which recorded its worst first-half performance this year since 1973. Safe haven assets like treasuries and the greenback have also suffered as markets price in tax cuts and spending hikes from Trump's recently passed "Big Beautiful Bill", which nonpartisan analysts predict will add more than $3 trillion to the country's $36.2 trillion debt. A weaker dollar tends to boost returns from foreign-currency assets. BlackRock recorded a positive foreign exchange impact of $171.52 billion in the quarter, compared with a $35.45 billion downward revision in the year-ago period. Its performance fees fell 42.7% to $94 million in the reported period, after falling nearly 71% in the first quarter. The low-cost investments juggernaut has been pivoting towards private markets, which provide higher margin revenue compared to those from ETFs, where BlackRock faces intense competition and fee compression as the market matures. Private markets saw inflows of $6.82 billion in the quarter. The New York-based firm said at its investor day last month that its private markets and technology businesses would make up 30% or more of its total revenue by 2030, up from 15% in 2024. As part of this push, BlackRock last month unveiled plans to include private assets in its retirement plans, which account for more than half of the money the company manages. Technology services revenue rose 26.3% to $499 million, reflecting the first full quarter of data provider Preqin, which BlackRock bought in a $3.2 billion deal last year. The deal closed on March 3, 2025. Its total revenue - most of which is earned as a percentage of assets under management - rose to $5.42 billion from $4.81 billion a year ago. BlackRock's total expenses rose to $3.69 billion from $3.01 billion last year. Adjusted profit came in at $1.88 billion, or $12.05 per share, for the three months ended June 30, up from $1.55 billion, or $10.36 per share, a year earlier.
Daily Mirror
an hour ago
- Daily Mirror
Elon Musk's Tesla on trial after car on autopilot kills woman as she's stargazing
Lawyers for the plaintiff argue the autopilot should have warned the driver and braked when his Model S blew through flashing red lights, a stop sign, and a T-intersection at nearly 70 miles an hour A rare trial against Elon Musk's car company has begun over the death of a university student after a runaway Tesla sent her flying 75 feet through the air. Lawyers for the plaintiff argue that the EV maker's driver-assistance feature, Autopilot, should have warned the driver and braked when his Model S blew through flashing red lights, a stop sign, and a T-intersection at nearly 70 miles an hour in the April 2019 crash. Tesla lays the blame solely on the driver, who was reaching for a dropped mobile phone. 'The evidence clearly shows that this crash had nothing to do with Tesla's Autopilot technology,' Tesla said in a statement. 'Instead, like so many unfortunate accidents since cellphones were invented, this was caused by a distracted driver.' The driver, George McGee, was sued separately by the plaintiffs. That case was settled. Now, a jury trial case is being heard in Florida over whether Musk's car company is liable. A judgment against Tesla could be especially damaging as the company works to convince the public its self-driving technology is safe during a planned rollout of hundreds of thousands of Tesla robotaxis on US roads by the end of next year. A jury trial is rare for the company, whose suits over crashes are often dismissed or settled, and this one is rarer yet because a judge recently ruled that the family of Naibel Benavides Leon can argue for punitive damages. The judge, Beth Bloom of the US District Court for the Southern District of Florida, issued a partial summary judgment last month, throwing out charges of defective manufacturing and negligent misrepresentation against Tesla. But she also ruled that plaintiffs could argue other claims that would make the company liable and ask for punitive damages, which could prove costly. 'A reasonable jury could find that Tesla acted in reckless disregard of human life for the sake of developing their product and maximising profit,' Bloom said in a filing. The 2021 lawsuit alleges the driver relied on Autopilot to reduce speed or come to a stop when it detected objects in its way, including a parked SUV that Benavides and her boyfriend, Dillon Angulo, had gotten out of near Key West, Florida, to look up at the sky. The Tesla rammed the Tahoe at high speed, causing it to rotate and slam into Benavides, tossing her into a wooded area and killing her. In legal documents, Tesla denied nearly all of the lawsuit's allegations and said it expects drivers to follow vehicle warnings and instructions in the owner's manual and comply with laws. Tesla's manuals warn owners that its cars cannot drive themselves and that they must always be ready to intervene. Plaintiff lawyers argue that Tesla should have geofenced its Autopilot so it only worked on the big roads it was designed for, and drivers couldn't use it on smaller roads, such as the rural one where Benavides was killed. They also say data and video evidence show the Autopilot did detect the Tahoe but then failed to warn the driver, as they claim it should have done. Tesla has since improved its driver-assistance and partial self-driving features, but still faces lawsuits and investigations over what critics say is a gap between its exaggerated depictions of how well they work and what they can actually do. Government auto safety regulators recalled 2.3 million Teslas in 2023 because Autopilot failed to sufficiently alert drivers if they weren't paying attention to the road. They then investigated the company last year for saying it fixed the problem, though it was unclear if it actually did. Musk has also continued to make public comments suggesting Tesla's 'Full Self-Driving' technology, a more advanced version of Autopilot, allows cars to drive themselves despite regulators' warnings not to. They say it could lead to overreliance on the systems, crashes and deaths. That technology has been involved in three fatal crashes and is under investigation of its ability to see in low-visibility conditions such as sunlight glare or fog. Tesla is promising a much more robust self-driving technology to power its robotaxis with no one behind the wheel. A taxi test run in Austin, Texas, appears to have gone mostly well, though there were scattered problems, such as when one cab went down the opposing lane.
BreakingNews.ie
an hour ago
- BreakingNews.ie
AI intensifying competition in Irish jobs market, survey finds
Irish workers say artificial intelligence (AI) is both creating opportunities and intensifying competition in the jobs market, according to a corporate survey. The hiring software firm's survey indicated that half of Irish workers fear for their jobs amid economic uncertainty and nearly two in three are struggling to navigate the jobs market – with AI intensifying competition. Advertisement Hiring company Greenhouse conducted a survey of 2,200 candidates, including 169 Irish-based workers along with workers from the UK and the US. Seventy-three per cent of Irish workers indicated they are using AI when looking for a new job, mainly for interview preparation (42 per cent), analysing job ads (28 per cent) and generating work samples (25 per cent). A further 54 per cent said AI is making job hunting harder by increasing skill standards and intensifying competition, while 41 per cent said it created and helped uncover new opportunities. The survey also indicated there is a lack of clarity on whether AI can be used when applying for jobs, with 82 per cent of workers stating that employers provided little or no guidance on using AI in interviews. Advertisement Nearly half of Irish job seekers said they feel insecure in their current role, while 42 per cent said the job market is very competitive. Chief executive of Greenhouse, Daniel Chait, said hiring is 'stuck in an AI doom loop'. 'As this technology advances, it makes it easier than ever to apply, flooding the system with noise,' he said. 'With 25 per cent of Gen Z saying AI has made it harder for them to stand out, candidates entering the market are up against more applications, more automation, and less clarity.' Advertisement Business Dublin TikTok worker replaced by AI loses claim fo... Read More The survey also indicated that 49 per cent of Irish job seekers said they had been asked inappropriate or biased questions during job application processes. The most common of these was about health or disability status (21 per cent), parental responsibilities (20 per cent), and age (18 per cent). A further 69 per cent said they had removed older experience from their CVs to try and avoid age-based assumptions, according to the survey.
