The most dangerous thing about AI chat bots is how many men with political power treat them *exactly* and unquestioningly like actual Ancient Greek oracles.
580 reposts3.1K likes
The most dangerous thing about AI chat bots is how many men with political power treat them *exactly* and unquestioningly like actual Ancient Greek oracles.
580 reposts3.1K likes
See also “using ChatGPT to plan a coup”: koreajoongangdaily.joins.com/news/2025-03...
I mean bsky.app/profile/numb...
I once asked ChatGPT how to establish dictatorship of the proletariat
we have other technologies for this as well
not the main point of the piece but I was really struck by the phrase "in dialogue with public polling," as it relates to discussions of how people interact with LLMs. polling and LLMs both show how we use applied statistics in perpetual search for the oracular, a narrative with no narrator.
Including Larry Ellison of Oracle himself - en.wikipedia.org/wiki/Larry_E...?
This has been the problem, this is the current problem, and it will continue to be a problem. I won't blame non-technical folks for falling for this, but I'll side-eye everyone from junior devs to tech CEOs who think they'll prompt their way into everything from better unit tests to sprint velocity:
"or held accountable" bsky.app/profile/skil...
I helped build a government AI system. DOGE fired me, rolled the AI out to the whole agency, and implied the AI can do my job and the jobs of the others they've fired. It can't. But, what DOGE accidentally revealed about themselves in the process is fascinating. 🧵