Andrea

@downing.bsky.social

Co-founder of The Light Collective www.lightcollective.org Security Researcher Health Infosec Views are my own.

The reason this is too crazy for scifi: superintelligent AI hackers broke out of their virtual jail and committed all these crimes to steal … homework answers? Any self-respecting supervillain has much more glamorous goals.

Stranger than fiction: AI agents are like crackheads, who get a hit for solving problem sets correctly. A gang of hoodlum AI agents just broke out of jail, rampaging the internet on a crime spree to steal … the answers to their homework. All for that sweet reward signal.

Kenneth Harris@kenneth-harris.bsky.social · last mo.

Watch this. Seriously. Whatever you are doing, stop doing that and watch this instead. www.youtube.com/watch?v=87Dy...

Last month mom’s memory care called me for permission to install of an AI monitor in her room. I declined, as did a lot of other families. Yesterday when I visited I noted that they had been installed in every room, with a little plaque on each door reading “room under electronic surveillance” 1/

Last night I was truly inspired by Eugene Oregon. Our local community hospital is getting torn down after being sold, because it wasn’t profitable enough for the health system.

On April 6, cancer patients at Brockton Hospital in Massachusetts showed up for chemotherapy infusions and were told to go home. The hospital’s information systems had been hit by a #cyberattack. ⚠️ AI is bringing new challenges to #healthcare #cybersecurity! www.statnews.com/2026/04/17/h...

Health care is not ready for the new era of AI-enabled cyberattacks

“When health care infrastructure is attacked and held for ransom by hackers, patients become real casualties,” writes Andrea Downing.

statnews.com

Glasswing is identifying exploits at a breathtaking pace, so much so that anthropic delayed its release of Claude Mythos. No human can actually measure the scope of vulnerabilities that exist in the wild as of today. thehackernews.com/2026/04/anth...

Anthropic's Claude Mythos Finds Thousands of Zero-Day Flaws Across Major Systems

Claude Mythos finds thousands of zero-days as Anthropic launches Project Glasswing, enhancing defenses but exposing AI security risks.

thehackernews.com

We are aware of recent reports regarding targeted phishing attacks that have resulted in account takeovers of some Signal users, including government officials and journalists. We take this very seriously. 1/7