It is interesting to me how controversial this post is. It has the highest upvotes, and most disagreeing comments, of anything I've typed up on HN.
I'll gladly admit I think what these companies are doing is unethical, and I'm sure that biases my thinking toward skepticism.
That said, there remains way too much that is hidden to be able to effectively evaluate what is going on. You have the perfect storm:
- AI companies do not share their custom internal harnesses.
- AI companies do not share their custom internal training data.
- AI companies do not share how much compute they allocate to trying to solve problems of this nature.
- AI companies are primarily marketing their models to investors as human-replacing rather than human-augmenting.
- AI companies are under enormous financial pressure to make their business work.
The last two points incentivize them to find these types of "first proof" successes as aggressively as they can, and I'm sure they've thrown the whole book at it.
Is it likely that they literally had a mathematician discover this, put it into the training data, and then prompted it out? Of course not.
But it would make a world of difference--in evaluating the impressiveness of this discovery and LLM capabilities in general--if we were to know the extent to which the training data crosses over this problem, the harness with which this was ran, and how much compute was spent.
Until they bring more transparency to the whole process--something which some of the mathematicians commenting on this even asked for--I will personally take discoveries of this nature with a good dose of salt.
That makes sense to me, overall it sounds like an 80/20 situation: 80% of drivers are side hustlers and will be harmed when they lose their side hustle, 20% will get a slight pay bump and benefits. The overall economic benefit for California will be a wash (and the overall result for the average driver, reduced income).
The real underlying problem is that we've been papering over our underemployment crisis with these low-skill, low wage jobs. You can't turn them into viable careers by legislative fiat.
> The real underlying problem is that we've been papering over our underemployment crisis with these low-skill, low wage jobs. You can't turn them into viable careers by legislative fiat.
It's worse. We have been papering over the lack of any social security net and lack of affordable healthcare by forcing these things onto an employer-employee relationship.
If you try to rethink the system from the ground up, there is no logical reason health insurance needs to be tied to employment. Employment law should be about removing highly coercive situations, but other than that, adults should be allowed to provide a service in return of a payment without getting embroiled in byzantine regulatory requirements.
I find the paternalism of AB5 to be a little bit problematic. It really assumes that the drivers are too stupid to do their own accounting and figure out if they are making money or not.
Pretty sure most drivers wouldn't be driving for $9.21 if a job at the local ice cream parlor could get them almost double on the spot. Yet you still see folks driving 30 hrs per week for years.
> It really assumes that the drivers are too stupid to do their own accounting and figure out if they are making money or not.
This isn't as easy as it sounds. Much of their income will go towards paying vehicle maintenance costs, and those are not obvious from the start. Your accounting can show you making decent wages for six months and then suddenly you need new tires and recalculating your wage will show you aren't making a ton of money. At which point many of them do quit and work somewhere else.
It's no more paternalistic than any minimum wage law. Society has decided that jobs can't pay below a certain threshold, the law just makes that apply to gig workers now too.
Fair, but that's also not any more paternalistic than all the existing laws which define which kind of work can be classified as a contractor and which must be classified as an employee.
This is my take too - that said, I think this is a structural problem across low-wage jobs. I don't think driving for Uber is inherently worse than working for Walmart or Amazon.
There needs to be greater orientation around skill acquisition that helps drivers, package handlers, etc. move up the skill stack.
>The real underlying problem is that we've been papering over our underemployment crisis with these low-skill, low wage jobs. You can't turn them into viable careers by legislative fiat.
I don't know anything about the business or cultural elements of Udacity, but they have such high quality of courses that this makes me genuinely sad. I hope they can figure out a business model - even if it leans B2B vs. B2C.
Congrats to the OpenAI team for one of the most significant breakthrough discoveries in AI history.
reply