Tags: artificial intelligenc

Bedrock Robotics raised $80 million to retrofit bulldozers and excavators with AI—not to cut jobs, but to solve a massive labor shortage and make dangerous work safer, it says.

OpenAI’s new ChatGPT Agent can do complex tasks on behalf of users, but it comes with potentially greater risks.

A new paper from 40+ AI safety researchers proposes monitoring the "thoughts" of AI models before they act. Experts warn this safety tool could become a surveillance weapon.

Google adds AI tools to Search that make calls and answer complex queries on users’ behalf, but experts warn about the hidden cost.

xAI‘s new Companions feature brings anime waifus to Grok subscribers, sparking everything from marriage proposals to existential dread about humanity‘s future.

The U.S. Department of Defense has awarded xAI a lucrative deal just days after the Grok chatbot spewed antisemitism, sparking concern.

While Elon Musk‘s $300 “Heavy” model puts up record-breaking benchmark scores, the basic Grok 4 struggles to keep up with competitors.

Elon Musk‘s new model crunched Polymarket odds and picked the Dodgers, but ChatGPT and DeepSeek favor the Tigers as a better value for bettors.

Coinbase has teamed with Perplexity AI to bring real-time crypto data into AI-powered search—and Brian Armstrong sees major benefits.

Studios agree to "strike suspension language" preventing AI scabs and substantial pay increases through 2027.

xAI’s latest AI model aims to challenge OpenAI, even as the team quietly addresses past missteps this week.

A new study of 15 million biomedical papers found telltale language patterns linked to ChatGPT, suggesting that AI tools are reshaping scientific writing far more than expected.

Grok‘s Nazi-sympathizing responses sparked backlash this week, but the fix reveals how a single prompt instruction can shape an AI‘s entire political worldview.

Grok maker xAI quietly updated its chatbot to assume all media is biased, relying on X, a platform known for misinformation.

The latest xAI model promises specialized coding capabilities, benchmark-beating reasoning, and something GPT-5 doesn‘t have: availability.

After ChatGPT falsely claimed that a feature existed at a music tabs website, Soundslice built it to meet user demand.

The executive‘s departure signals intensifying competition for elite talent and deepens concerns over Apple‘s generative AI strategy.

A developer tested 17 language models and discovered they‘re overwhelmingly associated with the Ravenclaw team, with one notable exception.

OpenAI’s new flagship model GPT-5 is nearly here. Industry insiders say it could be a game-changer—or just another overhyped upgrade.