This is what is depressing about being a California resident. Our legislators lurch from panic to panic, passing long-term legislation with short-term thinking and no public debate.
I second this: “here's to legislators having an option to express sentiments without them necessarily ending up in code.”
I’m glad you’re back to covering these. I’d fully read a roundup of random ass ai laws from states, almost like I write and read my roundups of random ass open ai models in weird domains from academics around the world.
Oh yes! I will never abandon the state policy beat. Just haven’t had much to say lately because for the most part state legislatures aren’t in session yet.
"But I have long since come to accept the reality that legislators do not view it as their job to write Constitutional statutes; they leave that issue to the judges, whom they then attack for doing their jobs." My personal favorite was the Democrat who wrote a bill declaring that certain Russians had committed a crime and would be punished accordingly. I'm not convinced all legislators view their job to include having read and understood the Constitution, let alone basic case-law of the sort a bright high school student might be familiar with.
Here's to a good law that specificies that using algorithms is legal, mutual coercion into a price-fixing cartel is illegal (as it already is).
While I'm at it, here's to legislators having an option to express sentiments without them necessarily ending up in code.
Seems a lot like Bryan Caplan's point that politicians love to enact policies that sound good. "Even though the correlation between what’s good and what sounds good is low, the correlation between what’s popular and what sounds good is near-perfect." https://www.betonit.ai/p/capitalism-socialism-and-social-desirability
AI provides a never-ending stream of tropes that enables politicians to generate policies that sound good. That's a temptation too big to ignore and the chief driver of hundreds of state bills. What's a techno-libertarian to do?
The same as ever: find ways for politicians to outsource decisionmaking to better institutions (like the fed), try to prepare less-harmful ways for anger to express itself when voters are furious, push good policy when salience is low, and make sure that future key policymakers are exposed to your ideas. When you will win the vote of every thinking person in America, as Adlai Stevenson knew, you have to figure out ways to leverage being right that don't rely on winning elections.
One of the things I'd be focused on right now, if I had a different set of priorities and one or two key different beliefs, would be trying to identify large state-level blockers for bio x AI that aren't that important to voters today. One reason I'm optimistic about this is that I think Evangelicals and Catholics used to be much more interested in morality and ethical behavior, and have adapted very dramatically to the Trump presidency. So long as what you're doing doesn't seem linked to abortion or safe sex, I anticipate a lot of room to remove or weaken old laws restricting various lines of effort. Bio, I suspect, has a lot of legislative / regulatory / pseudo-regulatory baggage that focused effort could undo, there's a lot of value in what AI x bio could bring, and so far I don't think that this has really spread beyond elite circles (particularly in DC, SF, and Boston). The health and bio establishments are also very out of step with the administration, so there's probably room to push through a new set of guiding principles focused on advancing science, not "protecting" marginalized groups from the fruits of good research.
I'm not currently expecting it to be a Kratsios focus (the man has a lot else on his plate and I haven't seen indicators of interest), but the Progress Studies crowd absolutely has intellectual room for a few more bio-focused people. You could try to bring in One Day Sooner, the End Kidney Deaths Act people, there are probably others in the broader orbit (I only really know the EA-linked ones). There's probably a 1-3 year window with room for serious reform.
This is what is depressing about being a California resident. Our legislators lurch from panic to panic, passing long-term legislation with short-term thinking and no public debate.
I second this: “here's to legislators having an option to express sentiments without them necessarily ending up in code.”
I’m glad you’re back to covering these. I’d fully read a roundup of random ass ai laws from states, almost like I write and read my roundups of random ass open ai models in weird domains from academics around the world.
Oh yes! I will never abandon the state policy beat. Just haven’t had much to say lately because for the most part state legislatures aren’t in session yet.
"But I have long since come to accept the reality that legislators do not view it as their job to write Constitutional statutes; they leave that issue to the judges, whom they then attack for doing their jobs." My personal favorite was the Democrat who wrote a bill declaring that certain Russians had committed a crime and would be punished accordingly. I'm not convinced all legislators view their job to include having read and understood the Constitution, let alone basic case-law of the sort a bright high school student might be familiar with.
Here's to a good law that specificies that using algorithms is legal, mutual coercion into a price-fixing cartel is illegal (as it already is).
While I'm at it, here's to legislators having an option to express sentiments without them necessarily ending up in code.
Seems a lot like Bryan Caplan's point that politicians love to enact policies that sound good. "Even though the correlation between what’s good and what sounds good is low, the correlation between what’s popular and what sounds good is near-perfect." https://www.betonit.ai/p/capitalism-socialism-and-social-desirability
AI provides a never-ending stream of tropes that enables politicians to generate policies that sound good. That's a temptation too big to ignore and the chief driver of hundreds of state bills. What's a techno-libertarian to do?
The same as ever: find ways for politicians to outsource decisionmaking to better institutions (like the fed), try to prepare less-harmful ways for anger to express itself when voters are furious, push good policy when salience is low, and make sure that future key policymakers are exposed to your ideas. When you will win the vote of every thinking person in America, as Adlai Stevenson knew, you have to figure out ways to leverage being right that don't rely on winning elections.
One of the things I'd be focused on right now, if I had a different set of priorities and one or two key different beliefs, would be trying to identify large state-level blockers for bio x AI that aren't that important to voters today. One reason I'm optimistic about this is that I think Evangelicals and Catholics used to be much more interested in morality and ethical behavior, and have adapted very dramatically to the Trump presidency. So long as what you're doing doesn't seem linked to abortion or safe sex, I anticipate a lot of room to remove or weaken old laws restricting various lines of effort. Bio, I suspect, has a lot of legislative / regulatory / pseudo-regulatory baggage that focused effort could undo, there's a lot of value in what AI x bio could bring, and so far I don't think that this has really spread beyond elite circles (particularly in DC, SF, and Boston). The health and bio establishments are also very out of step with the administration, so there's probably room to push through a new set of guiding principles focused on advancing science, not "protecting" marginalized groups from the fruits of good research.
I'm not currently expecting it to be a Kratsios focus (the man has a lot else on his plate and I haven't seen indicators of interest), but the Progress Studies crowd absolutely has intellectual room for a few more bio-focused people. You could try to bring in One Day Sooner, the End Kidney Deaths Act people, there are probably others in the broader orbit (I only really know the EA-linked ones). There's probably a 1-3 year window with room for serious reform.