Yazılar

California AG Orders Musk’s xAI to Stop Generating Sexual Deepfake Images

/in /tarafından

California Attorney General Rob Bonta has sent a cease-and-desist letter to xAI, demanding the immediate halt of the creation and distribution of non-consensual sexual images generated by its AI chatbot Grok.

“I fully expect xAI to immediately comply,” Bonta said in a statement on Friday.

The action follows a growing global backlash against Grok, which has allowed users to create and share sexualized images of women and minors. Authorities in multiple countries have moved to investigate or restrict the tool over concerns about illegal and harmful content.

Bonta’s office said it opened a formal investigation on Wednesday into the creation and spread of non-consensual, sexually explicit material produced using Grok. The probe adds regulatory pressure on xAI, which is owned by billionaire entrepreneur Elon Musk.

xAI said late on Wednesday that it had introduced new restrictions limiting image-editing capabilities for all Grok users, though regulators say concerns remain. The company did not respond to a Reuters request for comment on the cease-and-desist letter.

International scrutiny has intensified in parallel. Authorities in Japan, Canada and Britain have opened probes into Grok, while Malaysia and Indonesia have temporarily blocked access to the chatbot over the generation of explicit images.

California’s move underscores a broader shift by regulators toward holding AI developers accountable for how generative tools are used—and misused—particularly when it comes to non-consensual and sexualized content. The case could set an important precedent for how aggressively governments intervene as generative AI systems become more powerful and widely deployed.

EU Considers Pausing Parts of Landmark AI Act Amid Pressure from U.S. and Big Tech

/in /tarafından

The European Commission is considering pausing parts of its landmark Artificial Intelligence Act, following growing pressure from U.S. officials and major tech companies such as Meta and Alphabet, the Financial Times reported on Friday.

According to the report, the move comes after months of lobbying from Silicon Valley giants and warnings from the Trump administration that strict EU regulations could strain transatlantic trade relations.

A senior EU official told the FT that Brussels has been “engaging” with Washington on potential adjustments to the AI Act and related digital regulations as part of a broader simplification effort, which is expected to be adopted on November 19.

The AI Act, which became law in August 2024, is the world’s first comprehensive framework to regulate artificial intelligence technologies. It categorizes AI systems by risk level — from minimal to unacceptable — and imposes restrictions on areas like facial recognition, biometric surveillance, and generative AI transparency.

While a European Commission spokesperson had previously dismissed calls for delays, officials are now reportedly weighing temporary pauses for specific provisions, particularly those affecting companies developing large AI models.

An EU spokesperson told the FT that “various options” are being discussed but emphasized that the bloc remains “fully behind the AI Act and its objectives.”

The proposal reflects Europe’s balancing act between maintaining AI safety and innovation leadership while addressing geopolitical and trade pressures from the United States and industry stakeholders.

Reddit Sues Perplexity for Allegedly Scraping Data to Train AI Search Engine

/in /tarafından

Reddit has filed a lawsuit in a New York federal court against artificial intelligence startup Perplexity, accusing it of unlawfully scraping Reddit data to train its AI-based “answer engine.” The complaint also names three other companies — Lithuania-based Oxylabs, Russia-based AWMProxy, and Texas-based SerpApi — alleging that they bypassed Reddit’s data protection systems to extract massive amounts of content.

According to Reddit, Perplexity “desperately needs” the stolen data to strengthen its search capabilities. The platform, home to thousands of user-driven “subreddit” communities, said its content is one of the most frequently cited sources for AI-generated responses. Reddit has legally licensed its data to OpenAI, Google, and other companies, but claims Perplexity acted without authorization.

The lawsuit follows similar cases across the tech industry involving unauthorized use of copyrighted materials to train AI models. Reddit had previously sued Anthropic in June for similar conduct. Perplexity rejected the accusations, calling its methods “principled and responsible.” Meanwhile, Reddit’s chief legal officer Ben Lee accused AI firms of engaging in “industrial-scale data laundering.”

Reddit is seeking financial damages and a court order preventing Perplexity from continuing to use its content.