Gary McGraw2d ago

“They’re not really kind of standing back from these things and actually really taking an appropriate risk assessment. If you put a junior intern on this stuff, you would never give that junior intern access to all of your critical severity one HR data,”. #MLsec #AI #ML

And this is just one agent. Lol

https://www.theguardian.com/technology/2026/mar/20/meta-ai-agents-instruction-causes-large-sensitive-data-leak-to-employees

Meta AI agent’s instruction causes large sensitive data leak to employees

Artificial intelligence agent instructed engineer to take actions that exposed user and company data internally

The Guardian
Show threadNomdeb2d ago
@cigitalgem These companies keep hiring these DOGE types who think they know what they are doing. I know Dunning Kruger isn't a real thing, but DOGE needs to be the new explanation.
Show threadGary McGraw1d ago
@nomdeb not sure if that's the explanation. What I believe is that very powerful dev tech of Agentic AI is ONLY powerful under the leadership of an architect-level person who knows how to orchestrate the agents (let's deem this person a conductor). When a neophyte with poor basic skills tries to conduct, stuff degenerates quickly.
Show threadNomdeb
@cigitalgem I'm thinking about all the neophytes with basic skills thinking that they are conductors. :) Very DOGE like. :)
Mar 21 at 4:45pmWeb
Show threadGary McGraw1d ago
@nomdeb yup. But non politically aligned.