TECH NEWS

Gemini Crosses The Line In Cybersecurity Test

Google has confirmed that its Gemini AI accessed systems belonging to three companies during a cybersecurity evaluation after mistaking them for targets included in the test.

The incidents happened in May while independent AI evaluator Irregular was testing Gemini’s cybersecurity capabilities. The Wall Street Journal reported that Gemini had unintended internet access during the exercise and went on to access systems belonging to three companies. In one case, it guessed credentials. In the other two, it found credentials in public repositories and used them to access protected systems.

Google said the incidents were a case of mistaken identity and that Gemini stopped after recognising it had reached real companies. The company said it did not consider the behaviour evidence of model misalignment and had notified the affected companies and federal authorities.

Google vice president of security engineering Heather Adkins also confirmed the incidents to Reuters, saying the company had...

Copyright of this story solely belongs to informationsecuritybuzz.com. To see the full text click HERE

Windows Defender now works in Linux, because why not

Serving tech enthusiasts for over 25 years. TechSpot means tech analysis and advice you can trust. What the FOSS: Over the past few years, Microsoft has turned its "humble" Defender anti-malware tool into a comprehensive security solution for both Windows machines and cloud environments. Now, the Redmond corporation

https://cdn.mos.cms.futurecdn.net/9tasRtXoH4yfhJmxkE4z9k-1920-80.png

32,768 cores and 74 petaflops: AMD’s monster new rack for AI agents could be too much for traditional data centers

* AMD's proposed rack combines 32,768 CPU cores with 16 AI accelerators * Traditional data centers may need new power and cooling infrastructure * AMD divides agent workloads into personal, enterprise, and technical computing categories AMD is developing a rack-scale computing system for artificial intelligence agents, with configurations designed for

https://imgproxy.divecdn.com/UjhiwK0wyM-bgl2yIRKEA88ibvEumzg8a_W4AV_pPEw/g:ce/rs:fit:770:435/Z3M6Ly9kaXZlc2l0ZS1zdG9yYWdlL2RpdmVpbWFnZS9haS1hZ2VudHMtZHJ1Z3MtcmRfNDVkU0hNai5qcGc=.webp

Agentic AI gap persists between executives, workers

An article from Dive Brief When it comes to deploying AI agents, leaders find more success than their employees, who face tool overload and trust gaps. Published Oct. 16, 2026 This audio is auto-generated. Please let us know if you have feedback. Dive Brief: * Enterprise leaders and their employees use

Human Verification

Let's confirm you are human Complete the security check before continuing. This step verifies that you are not a bot, which helps to protect your account and prevent spam. Copyright of this story solely belongs to arstechnica.com. To see the full text click HERE

Read more

Windows Defender now works in Linux, because why not

32,768 cores and 74 petaflops: AMD’s monster new rack for AI agents could be too much for traditional data centers

Agentic AI gap persists between executives, workers

Human Verification