I've been making that point many times:
1. Intelligence is not a scalar quantity. The space of problems is gigantic and, for a given amount of computing resources, any intelligent system will only excel at a tiny subset of them. Any intelligent system has blind spots, hence vulnerabilities.
2. Intelligence does not imply an ability to dominate humans. It depends on *what* type of intelligence.
3. Intelligence does not imply a *desire* to dominate. Even *within* the human species, intelligent people rarely want to dominate others. If they did, Einstein and Feynman would have been rich and powerful. They were neither (also true of many of my scientist colleagues). In fact, Geoff points out that, until the election of Biden, the most powerful man in the world was not particularly smart (weakening his own argument).
4. Geoff mentions the fact that gaining power is an instrumental subgoal of achieving any goal. But it's easy to design guardrail objectives to prevent bad things from happening. We already do this all the time by making laws, including for superhuman entities such as corporations and governments.
To me, the most thing about today's new CPI data is that inflation except for housing is 2.2% (right on the Fed's target) but housing inflation is 5.5% (nearly triple the Fed's target).
The housing shortage is the single largest contributor to inflation right now.
This is incredibly dangerous. It lays the path for centralized, device-level client side scanning.
From detecting 'scams' it's a short step to "detecting patterns commonly associated w/ seeking reproductive care" or "commonly associated w/ providing LGBTQ resources" or "commonly associated with tech worker whistleblowing."
Do not agree to have Google wiretap your calls. This is exceptionally dangerous from a ton of different perspectives, especially your civil liberties. It does not matter if it's on "on-device AI" -- that's got nothing to do with it's ability to call home, which it will do.
New York wouldn't build new housing so instead they made it hard to run airbnbs and banned new hotels and now they still have unaffordable housing but it also costs $1,000 a night for an airbnb
A serious “industrial policy” vis-a-vis China can’t just be tariffs, it needs to be complemented by *more* openness to trade and investment from Japan, Korea, Europe, Latin America, etc etc etc
https://t.co/3HPXhj8bEn
Telegram has launched a pretty intense campaign to malign Signal as insecure, with assistance from Elon Musk. The goal seems to be to get activists to switch away from encrypted Signal to mostly-unencrypted Telegram. I want to talk about this a bit. 1/
Previous policy efforts to stifle CS innovation have been much less dangerous because often we were fighting threatened industries who didn't have credibility in the spaces they were attacking (e.g. MPAA, RIAA, telcos).
This wave is different. Big Tech is an oligopoly protecting itself. It has the credibility to convince lawmakers that its capture ploys are legitimate. And it is the largest economic sector by a wide margin, so resources to staff an existential battle are effectively unlimited.
Further, they've bought and paid their way into academia's complicity in some cases, and complacency in others. And have employed a non trivial portion of the tech workforce, garnering outsized leverage on sentiment.
This is a very different game. And we really need to push for little tech representation in policy decisions for some modicum of fairness and balance.
Extremely sobering read. China is building a full-on Silicon Valley style ecosystem of ultra fast moving & hypercompetitive EV makers, while the US auto industry focuses on who can make a gas powered truck that most resembles a Bradley fighting vehicle https://t.co/jDZknFiVUv
This is the group behind SB 1047. Seriously, we need to stop the insanity.
Extinction from AI is science fiction and it's being used to justify terrible legislation in Ca.
We desperately need more sensible voices at the table.
Oh good god. DRM for models. It's like every bad, fucked up idea from the late 90's and early 2000's is being dredged up again for the AI safety wars.
https://t.co/2GLkhq9z4e