tfehring

11 ms·

91.
by tfehring 2y ago
The author is claiming that Bayesians vary along two axes: (1) whether they generally try to inform their priors with their knowledge or beliefs about the world, and (2) whether they iterate on the functional form of the model based on its
92.
by tfehring 2y ago
https://archive.ph/e401r
93.
by tfehring 3y ago
> * If an insurance company wanted to charge higher premiums, they would just do so.* That’s not an option in many cases. The first homeowner featured in the article is in rural Northern California. CA’s insurance regulator has been extr
94.
by tfehring 3y ago
There’s a saying in the insurance industry, there’s no such thing as a bad risk, only insufficient premium. Natural catastrophe risk is definitely increasing, but the insurance industry can handle that. The fundamental issue is that regulat
95.
by tfehring 3y ago
Yes, insurance companies should sometimes lose money on business that they expected to be profitable. But they shouldn’t be writing business that they expect to be unprofitable a priori . Many states and lines of business are firmly in the
96.
by tfehring 3y ago
It would be extremely poor form for founders to intentionally leave their startup at this stage to take another job. In all likelihood they were pushed out by the board. My guess is that the founders were all in on Pi, their conversational
97.
by tfehring 3y ago
I agree that there's still lots of ambiguity and Waymo's PR comment provides limited and one-sided detail. I stand by my previous comment as the most likely course of events, but we won't know for sure until video is released
98.
by tfehring 3y ago
Yeah, my reading was that the bike ran the stop sign and then attempted to make a left turn in front of the Waymo as it was going straight through the intersection.
99.
by tfehring 3y ago
You get the full $500k upfront. In exchange, YC gets 7% equity upfront (for the first $125k), plus a proportional share of equity in your next funding round (for the next $375k).
100.
by tfehring 3y ago
There’s not a fixed federal limit on loss ratio or profit for car insurance like there is for some types of health insurance. But they need to get approval for rate increases from state regulators for every state that they issue policies in
101.
by tfehring 3y ago
Long-term interest rates aren’t set by the Fed, they’re market rates based on the Treasury’s demand for borrowing and investors’ demand for purchasing US government debt. If there were enough demand for zero-interest US government debt to s
102.
by tfehring 3y ago
Netflix is one of the most data-driven companies there is. They know exactly how much customers care about those things, as measured in dollars of revenue, and they account for it in their product and content decisions. As a consumer, I wis
103.
by tfehring 3y ago
https://nitter.net/dhh/status/1743341929675493806
104.
by tfehring 3y ago
In the US, health insurers can only price based on age, location, and tobacco use. Setting health insurance premiums or denying coverage based on any health-related factors has been illegal for over a decade, and changing that would be tota
105.
by tfehring 3y ago
My impression is that it's pretty true of Microsoft in general, but not of the AI research teams I'm familiar with. As with any tech company of its size, there's lots of variation from team to team.
106.
by tfehring 3y ago
It's frustrating, but she's exposed to potentially massive personal liability for the board's actions. I'm surprised that she agreed to an interview at all - frankly, I probably wouldn't if I were in her position -
107.
by tfehring 3y ago
Dalton says basically the same thing in the video. Lightly edited excerpt from the YouTube transcript (starting around 11:28): What often happens - again, to go deep on this - is that the way they'll pitch the person is that they'
108.
by tfehring 3y ago
FYI you seem to be shadowbanned, I vouched for this comment but you should probably reach out to hn@ycombinator.com
109.
by tfehring 3y ago
My main concern is that a new Altman-led AI company would be less safety-focused than OpenAI. I think him returning to OpenAI would be better for AI safety, hard to say whether it would be better for AI progress though.
110.
by tfehring 3y ago
I think it was a good thing that, in hindsight, the leading AI research company had a strong enough safety focus that it could do something like this. But that’s only the case as long as OpenAI remains the leading AI research company going
111.
by tfehring 3y ago
I’m sure they had A/B tested the feature, but it’s hard to think of a metric they could measure in those tests that would reliably detect the alleged harm. I’m not sure they made the right choice in the absence of data, but I do believ
112.
by tfehring 3y ago
Yep, insurance is really at quantifying and transferring risk. Maybe the best thing about it is that it translates high, unlikely costs in the future into small, certain costs today, and the latter are far easier to respond to rationally an
113.
by tfehring 3y ago
> [Matt] Dorsey, a former police-department spokesperson, had the idea of adding fentanyl dealing to a list of crimes for which undocumented immigrants lose their sanctuary-city protections. "My hope was to incentivize a change to
114.
by tfehring 3y ago
Financially, yeah, you can cash out and get similar liquidity to what you'd get from an exit while keeping plenty of upside. But investors are only going to provide that option in the situation where they like you as a CXO and think yo
115.
by tfehring 3y ago
> The only way to avoid vendor lock in is to allow passkeys to be persisted unencrypted. Is this true? Naively I’d expect there to be a two-key solution that would allow Vendor A to transfer passkeys to Vendor B without requiring them
116.
by tfehring 3y ago
This is very close to what I’d do, but with the addition of enough office space for anyone who wants it, and less frequent “required” gatherings rather than once a month. It’s really counterintuitive, as suggested by all the other replies s
117.
by tfehring 3y ago
Based on the source data [0] I think the number they're reporting is just the total personal income earned in each state. In California's case, per-capita income actually increased slightly, but the population went down so total i
118.
by tfehring 3y ago
Diabetic mothers are significantly more likely than non-diabetic mothers to give birth to children with autism. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4181720/
119.
by tfehring 3y ago
That practice has been referred to recently as "shadow (insider) trading". It's an actively evolving area of case law, but the SEC's opinion is yes, see https://wp.nyu.edu/compliance_enforcement/2022
120.
by tfehring 3y ago
For reference, 90th percentile income globally is about $19,300 USD per year according to [0]. If you're in the top 10% by income, it's likely that you're also in the top 10% by emissions. [0] https://howrichami.gi

More ›