From crackdowns on discriminatory facial recognition to protections against deepfakes, 2024 was quietly an important year for AI regulation.

It didn’t take long to figure out how he wanted to play the trend.“Whenever you look at AI, it’s like, all the roads lead to Nvidia,” said MacGillivray, who’s spent thousands of dollars on shares this year from his home in Michigan.

Our tech columnist visits a doctor assisted by AI. He finds potential and worrisome questions about whether it’s accurate, biased or always useful in health care.

Meta is the world’s standard bearer for open-weight AI. In a fascinating case study in corporate strategy, while rivals like OpenAI and Google have kept their frontier models closed source and charged for their use, Meta has chosen to give its state-of-the-art Llama models away for free.

A new set of much more challenging evals has emerged in response, created by companies, nonprofits, and governments. Yet even on the most advanced evals, AI systems are making astonishing progress. In November,

Congress will try to spur AI growth and mitigate harms next year. But passing legislation will be an uphill battle

All of the major phone makers are at fault. Samsung opened the year with its Galaxy S24 launch in January, declaring “Galaxy AI is here” at a hockey arena-appropriate volume. To be sure, the devices it announced are good smartphones, and they run a blend of Samsung and Google’s Gemini Nano models on-device, but I wouldn’t call them AI smartphones.

Ishani Singh created Girls Rule AI after attending a computer science competition where she was the only girl.

AI offers business leaders the promise of higher efficiency and productivity. But there is risk in rushing to realize this potential. Nearly half of U.S. (47 percent) workers feel unprepared for its widespread adoption at their respective organizations according to recent SHRM research.

Agentic AI refers to systems that can autonomously take actions to solve problems or accomplish multi-step tasks on behalf of humans.

Databricks' vice president of AI, Naveen Rao, told Command Line that probably fewer than 1,000 people were capable of building frontier AI models.