AI Safety
Tech leaders are literally losing sleep over AI psychosis and "seemingly conscious" models
"Reports of delusions and unhealthy attachment keep rising - and this is not something confined to people already at risk of mental health issues."
Microsoft
Security
"An AI obedience problem": World's first LLM Scope Violation attack tricks Microsoft Copilot into handing over data
Zero-click bug requires "no specific user interaction and results in concrete cybersecurity damage", researchers allege.
Dark Web
"Security teams don’t need more noise": Microsoft Copilot now surfaces dark web threat intelligence
Bitsight teams up with Microsoft to monitor threats emerging from the dankest corners of the hidden web.
Opinion
Active Directory at 25: The growing role of automation in recovery
"Cybercriminals are well aware of the potential to breach network security..."
Microsoft
Microsoft gets into kitchenware, cooks up new "metasurface" tech
Redmond takes on the scourge of undercooked food and bursting eggs, empowering the world to make better microwave dinners.
Quantum
Microsoft launches quantum chip named after physicist whose death remains a mystery
"This breakthrough will allow us to create a truly meaningful quantum computer not in decades, but in years."
Microsoft
Microsoft admits new Phi-4 small language model is "tedious"
Waffly model delights in giving "long and elaborate" answers that can be just a little bit dull.
Elon Musk
Microsoft trains AI models on Elon Musk and George Soros conspiracy theories
But nothing about Bill Gates...
GenAI
Beyond large AI: The rise of lean language models
"LLM model size competition is intensifying... backwards."