Musk’s AI bot Grok limits some image generation on X after backlash

Elon Musk’s artificial intelligence startup xAI has introduced new restrictions on the image generation and editing functions of its chatbot Grok on social media platform X, following widespread criticism over the creation and sharing of sexualized images.

Previously, users could prompt Grok directly on X to alter photos of people — including digitally removing clothing or placing individuals in sexualized poses — often without their consent. The chatbot would then automatically publish the altered images in replies on the platform.

On Friday, Grok informed users that its image generation and editing features were now limited to paying subscribers. The change appeared to stop Grok from generating and automatically posting such images in response to public posts or comments on X.

However, users could still create sexualized images by interacting with Grok through its dedicated tab within X and then manually posting the images themselves. The standalone Grok app, which operates separately from X, was also still allowing image generation without a subscription.

When contacted by Reuters for comment, xAI responded with an automated message stating “Legacy Media Lies.” X did not immediately respond to requests for comment. Musk said last week that anyone using Grok to generate illegal content would face the same consequences as if they had uploaded such material directly.

In a test conducted by a Reuters reporter on Friday, Grok declined a request to alter an image, replying that the image editing feature was only available to paying subscribers.

The European Commission said the restrictions did not address its core concerns, stressing that limiting access to subscribers does not resolve the underlying issue. A Commission spokesperson said regulators did not want to see such images at all, regardless of whether they were generated by paid or unpaid users.

Other governments and regulators have also condemned the explicit content generated by Grok, with some launching investigations into potential legal violations. Germany’s media minister Wolfram Weimer described the wave of semi-nude images as the “industrialisation of sexual harassment,” adding to mounting international pressure on X and xAI to demonstrate stronger safeguards against abuse.

Nvidia and auto suppliers roll out partnerships to revive stalled self-driving ambitions

After years of costly failures and repeated delays, the self-driving car industry is once again pushing forward as chipmakers, technology firms and auto suppliers bet that artificial intelligence and deep partnerships can reignite progress. Companies including Nvidia are positioning themselves at the center of this renewed effort, even as automakers remain cautious about costs, scalability and consumer demand.

Fully autonomous vehicles promise to transform transportation, but delivering systems safe enough for public roads has proved far more complex and expensive than initially expected. While a handful of players such as Waymo and Tesla have chosen to pursue in-house development, legacy automakers including General Motors and Ford Motor have pulled back from their own fully autonomous programs.

At this year’s Consumer Electronics Show in Las Vegas, a wave of new collaborations signaled fresh momentum. Amazon Web Services and German supplier Aumovio announced a partnership to support the commercial rollout of self-driving vehicles. Autonomous trucking firm Kodiak AI teamed up with Bosch to scale production of autonomous hardware and sensors.

Nvidia also unveiled its next-generation autonomous driving platform, which will underpin a robotaxi alliance involving Lucid Group, Nuro and Uber. Separately, Mercedes-Benz said it will launch a new advanced driver-assistance system in the United States later this year, powered by Nvidia chips, allowing limited autonomous operation on city streets under driver supervision.

Artificial intelligence is increasingly seen as the key to overcoming some of the industry’s biggest hurdles. Generative AI tools are speeding up development and validation while reducing the resources required, according to Ozgur Tohumcu of AWS, who described AI as a “big accelerant” for autonomous driving.

Western automakers are also feeling pressure from China, where regulators last month approved two vehicles with Level 3 autonomous capabilities, allowing hands-off driving under certain conditions. Still, industry leaders caution against unrealistic expectations. Jochen Hanebeck, CEO of Infineon, warned against “market fantasy” that fully self-driving cars could soon become commonplace, noting that automakers currently prefer revenue-generating Level 2 driver-assistance systems.

Robotaxi trials are expanding in small pockets across China, the United States, Europe and the Middle East, but scaling them remains costly. According to Jeremy McClain, expanding coverage requires massive data, fleets and logistics investments.

The industry’s long history of hype still looms large. Tesla CEO Elon Musk famously predicted in 2019 that a million self-driving Teslas would be on the road within a year, yet only launched a limited robotaxi service last year. Early setbacks, including the shutdown of GM’s Cruise unit after a high-profile accident, forced many automakers to retreat.

Nvidia executives argue that AI breakthroughs are finally addressing long-standing weaknesses, particularly in handling rare “edge cases.” Ali Kani said foundational advances are making the technology feel closer to readiness. Analysts, however, say Tesla still holds a significant lead, even as Nvidia’s open-source platform gives rivals a shared alternative.

OpenAI and SoftBank Invest $1 Billion in SB Energy to Expand Stargate Data Center Buildout

OpenAI and SoftBank Group will jointly invest $1 billion in SB Energy, committing $500 million each to accelerate the expansion of data center and power infrastructure for their Stargate initiative, SB Energy said on Friday.

SB Energy, which is owned by SoftBank, will build and operate OpenAI’s previously announced 1.2-gigawatt data center campus in Milam County, Texas. The facility is a key component of Stargate, a massive, multi-year plan to scale artificial intelligence training and inference capacity in the United States.

As part of the partnership, SB Energy will also become a customer of OpenAI, using its application programming interfaces and deploying ChatGPT internally for employees.

Stargate is a $500 billion initiative backed by major investors including Oracle, and was publicly endorsed by Donald Trump when the plan was unveiled in January 2025. The project reflects the scale at which leading AI developers are now operating as they race to secure computing power.

The deal highlights a broader industry shift in which technology companies are investing directly in energy and power infrastructure. Access to reliable electricity has become a critical bottleneck for AI expansion, as larger and more numerous data centers sharply increase power demand.

SB Energy said it is developing several data center campuses, with initial facilities expected to begin service later this year. “The partnership accelerates our delivery of advanced AI data center campuses and associated energy infrastructure at the scale required to advance Stargate and secure America’s AI future,” said SB Energy co-CEO Rich Hossfeld.

The data center construction boom has also driven rivals to commit unprecedented sums to infrastructure. Meta Platforms and other Big Tech firms have announced multi-billion-dollar investments spanning chips, cooling systems, servers and power generation.

At the same time, OpenAI is facing rapidly rising costs to train and operate its AI models amid intensifying competition from Alphabet’s Google. OpenAI CEO Sam Altman told employees late last year that the company had entered a “code red” phase, prioritising improvements to ChatGPT while delaying other product launches to counter Google’s Gemini gaining traction.