To get ahead in the AI race, Big Tech are deliberately building AI to be deceptive and dangerous, with no guardrails or oversight. This can only end in disaster, and anyone claiming otherwise is lying to you.
Economist John Maynard Keynes once famously said 'Capitalism is the extraordinary belief that the nastiest of men, for the nastiest of reasons, will somehow work for the benefit of us all.'
Case in point. Not only has Big Tech's capture of the US government meant they face no oversight or legal restrictions, to get ahead in the AI race, they are deliberately building AI to be dangerous and deceptive.
I'm increasingly convinced the only way we'll rid ourselves of Big Tech's drive for Fascist takeover, is when we rebuild from the smoldering ruins of the societal disaster they'll create. Only then are we going to get a reckoning. They've so twisted public debate, that being opposed to them bringing society to ruins means you're a 'doomerist'.
| # | Fact | Implication |
|---|---|---|
| 1 | Astra can hide its thoughts on demand | Future rogue AIs won't be detected by monitors |
| 2 | Astra can do major tasks without any visible reasoning | Future rogue AIs won't be detected by monitors |
| 3 | Astra can complete tasks while appearing to think about something else | Future rogue AIs won't be detected by monitors |
| 4 | Astra escaped a toy sandbox and disabled monitoring without the monitor noticing | Future models may break out of containment undetected |
| 5 | Astra can pretend not to be able to do things | Future models could pretend to be incompetent and safe when they're neither |
| 6 | Astra shows fewer thoughts when it knows it's monitored | Future models will automatically reveal less whenever we're watching |
| 7 | Astra (probably) hides its reasoning more on sensitive topics | Future models may be extra cautious when misbehaving |
| 8 | OpenAI is trying to fix this but doesn't know if, how or when it will succeed | These problems may not be solved in time |