AI hacking tools like Mythos can be 'net positive' says top cyber official
The head of the National Cyber Security Centre says frontier AI tools can be a force for good - if kept out of the wrong hands.

The head of the National Cyber Security Centre says frontier AI tools can be a force for good - if kept out of the wrong hands.

Hesai Technology, a Shanghai-based lidar manufacturer, is under scrutiny for national security risks despite being blacklisted by the U.S. Department of Defense. Experts warn that Chinese lidar could create backdoors for Beijing to access sensitive data, while the company's CEO denies these claims. Hesai's technology is integrated into various autonomous systems, including those of Nvidia, Zoox, and Amazon.

A survey reveals 50% of employees feel they rely too much on AI, with many fearing it makes them less intelligent and harms career prospects. Experts advise using AI as a partner, not a crutch, by auditing AI use, doing difficult tasks manually first, and developing uniquely human skills.

Hospitals and businesses are increasingly renting robots, from medical supply movers like Moxi to humanoid entertainers and home helpers. This 'robotics-as-a-service' model offers flexibility, lower upfront costs, and access to the latest technology, addressing the rapid pace of innovation in the robotics industry.

Amazon's Leo satellite network is poised for initial service launch later this year after deploying over 390 satellites. The company aims to compete with SpaceX's Starlink, facing challenges from rocket capacity shortages and launch vehicle delays.

AI pioneer Yann LeCun is developing a new AI model, Joint Embedding Predictive Architecture (JEPA), at his startup AMI Labs. Unlike current LLMs like ChatGPT, JEPA aims to understand the physical world, enabling robots to perform complex tasks. The company has raised over $1bn from investors including Nvidia and Jeff Bezos's fund.

Meta is introducing a $19.99 monthly subscription for "Conversation Focus" on its AI glasses, limiting free usage to three hours. This move is part of broader "premium" feature experiments across Meta's platforms, sparking user concern despite other AI features remaining free.