Jimega36

"Access to frontier models no longer guarantees access to the inference regime that can reliably reproduce frontier model capabilities."

Switzerland has the system closest to the US in my experience, but with universal coverage.

As individual, you buy basic insurance costing roughly $500/month with a $2500 yearly deductible (LAMal/KVG). Applicants must be accepted even with pre-existing conditions. Post-deductible, you pay 10% co-insurance capped at roughly $800/year (no medical bankruptcy). Insurers operate basic plans as non-profit. They make their profit on optional supplemental insurance (private hospital rooms, extra benefits...) The state subsidizes basic premiums for low-income individuals. A doctor's visit is a $150 minimum, similar to the US.

What was most surprising compared to here in the US is employers pay $0 toward premiums. Funding is entirely decoupled from employment. You don't lose coverage or change plans when changing jobs.

On the negative, basic insurance does not have dental, so dental stuff is out-of-pocket and expensive.

In terms of stats, the US spends 18% of GDP on healthcare vs. 12% for Switzerland and the gain is lower stress too. Not perfect but maybe the best of both worlds.

Reach AGI to leapfrog whoever is behind. Burn everything to get there faster.

7314 (ChatGPT) 7,342 (Claude) 7492 (Gemini)

"In this post, we (a16z) will look at how AI will open new markets previously deemed too “small” to support a large VSaaS company. By increasing LTV (Life Time Value) per customer (through replacing labor with software) and reducing CAC (through leveraging AI-driven sales and marketing tools), we believe AI will open a plethora of new markets. Namely, the long tail of niche industries you were likely not thinking of before this post: chiropractors, dry-cleaning and laundry services, veterinary services, and more."

"The lower-power machine was much cheaper to manufacture and operate, more comfortable and less noisy for patients, and the final images after computational processing were as clear and detailed as those obtained by the high-power devices currently used in the clinical setting."

Surprised the exit happens at the relatively low price of $200M [1]. Klout had about 500 million users from what I understand. While monetisation might have been tough with deals, Klout got so much data. So much talk of 'big data' but it does not seem to be worth so much... or was there some founder fatigue at play here after almost 6 years of existence?

[1] http://www.crunchbase.com/company/klout

As a founder, it appears as if our startup's success or failure depends on us cofounders. This drives me to work insane hours as some sort of ego self rationalization tells me that more work will make it succeed.

In general I have seen the work/result correlation hold true. And yet, we all know how luck plays its role. I personally like to think some events are meant to happen or not. Especially when it goes bad - when my startup lost a client in the early days for example it was extremely tough... but we learned so much that the next bigger one went well. Would working more hours have changed anything, I doubt it, so there is more to it than work.

More than this, working many hours is sometimes just a way to reduce anxiety of success or failure, at least for me. What's tricky is understanding which hours not to do, how to make more of the ones I do, and being confident that cutting the hours won't affect the bigger picture output too much. Easier said than done.

Against big governments, nothing is stronger than the little men and women getting together.

If this is so bad (which I am not technically capable of understanding...), what shall we DO about it?

Signing the petition is probably not enough. Get people to have a minute of 'no internet' across the world? Similar to the Anti-SOPA movement? Suggestions welcome.

A fantastic simple idea. Video interaction is brilliant through gaming add-on. 3 comments:

1) Find a way to MAINTAIN gaming during the video (e.g. live comments adds of people in the room to the images?). That's because right now its fun to vote every 1:30minutes but then kind of boring to watch.. Althrough I stayed 3X more than initially planned ;-) This would allow maintaing people on site too since many people left/stopped voting after 1st video.

2) POINTS need to be prominent & STATUS linked: it's not about the money for me but the challenge & game. Many sites followed this model. When I win, the reward of points would need to incenvize me somehow. Make it big/prominent? What about leveraging STATUS with points too: status changes with results from say commoner to king to rock-star...

3) Consider Consumer Goods ads for site monetization: FMCG would pay to have their ads displayed (what about an ad of 10'' every few where people vote too???). I know that's old model but still there is something in the voting that could be a way for consumer research or brand 'interraction' of sort

Overall will be big, well done and keep it up!

Google = totally free lets not forget it => you give up some pricacy for service to be free in a way = not invasion of privacy but maybe price of service? Love to complain but if google does not make money, how can all thse cool apps be funded in practice. Advertising can be seen as privacy invasion too by the way. Money money money it's a ...

The way to monetize users should somehow be PERFORMANCE BASED rather than purely subscription based. Yes, subscription allows for great filtering of people..and yet, maybe it feels too costly at first (because the whole value/match is delayed to much later?). There must be a way to get money out of the HUGE satisfaction of finding the girl/guy when this happens - and this could complement (if not remove) the need for an entry cost to accessing the pool. For example, dance clubs make most money out of drinks (fun related) not from entrance fees (access to the pool of people). Maybe pay in order to go on a 2nd date? Pay every time you click on a profile (point system)? Pay a bonus when you find a match (how to prove...)? Do you know if this kind of stuff was ever tried or what do you think?

Using such a tool sounds like a vast lie from the companies benefiting from it. Isn't impersonation a crime? It would be fun to sue for massive amounts one of the fake users and then the company that generated would be liable:))) If ever proven so through.