Had the pleasure of having dinner with Vint one evening. Such an amazing person. So incredibly humble and one of the most genuine people Ive ever met. As an aside, I asked him directly about the Al Gore internet meme and he said point blank that if it weren't for Als work in Washington the Internet wouldn't exist in its current form.
But yeah, just an awesome guy all around. One of the few people I consider to be a source of inspiration for me.
The official "spiritual successor" seems to be Coincidence studio - their games in the genre being "Kaizen: A Factory Story" and the recent "U.V.S. Nirmana".
What's up with the AI narration at the beginning? Or is it just someone with an incredibly steady voice and AI cadence? It's uncanny and weird considering that this is a podcast hosted by actual humans!
However, it's one of those metrics that tends to be kind of meaningless. Vehicle safety team uses GPUs, so lets bill all GPU driver teams to the metric... that sort of thing
This reminds me of being told dogs don't feel emotions by someone who never owned one. Parrots most definitely can talk. Their language is extremely primitive but if you've ever been around a grey and it's owner for some time, they definitely talk to each other. The parrot will readily communicate observations and desires.
That's not how ethical disclosure works. Both parties should publish and we, the wider tech industry should see this as a good thing both for the hacker and the company that worked with them.
Not exactly. Infra will win the race. In this aspect, Google is miles ahead of the competition. Their DC solutions scale very well. Their only risk is that the hardware and low level software stack is EXTREMELY custom. They don't even fully leverage OCP. Having said that, this has never been a major problem for Google over their 20+ years of moving away from OTS parts.
But anyone with enough money can make infra. Maybe not at the scale of Google, but maybe that's not necessary (unless you have a continuous stream of fresh high-quality training data).
If making infra means designing their own silicon to target only inference instead of more general GPUs I can agree with you, otherwise the long-term success is based on how cheap they can run the infra compared to competitors.
Depending on Nvidia for your inference means you'll be price gouged for it, Nvidia has a golden goose for now and will milk it as much as possible.
I don't see how a company without optimised hardware can win in the long run.
The silicon can be very generic. I don't see why prices of "tensor" computation units can't go down if the world sees the value in them, just like how it happened with CPUs.
But yeah, just an awesome guy all around. One of the few people I consider to be a source of inspiration for me.