broken_clock

IDK why you needed AI to write the little blurbs, it's not really a lot of text.

Like if you feel it's not important enough to write yourself, just don't put anything there?

How is China (or any other country) any different except the same thing shifted upward?

https://en.wikipedia.org/wiki/List_of_administrative_divisio...

> If you elected a businessman or a lawyer, he/she'd tell you that Canadian exports to the US are 2x the imports from the US,

An honest one wouldn't, because it's categorically untrue.

"U.S. exports were $441 billion, while imports were $482 billion, resulting in a United States $41 billion trade deficit with Canada."

This doesn't even count the multinational tech corporations providing services in Canada through a local subsidiary, something that is not counted in balance of payments.

Can you share a source/article?

Aren't there already a ton of startups doing finetunes for their local niche? Many aren't even "AI" companies - it's pretty easy to slap a finetune together if you enough data.

If you mean developing a model from scratch just for your niche - the bitter lesson is that scale is everything and that a finetune from an internet-scale model will outperform you easily.

What tools are you using to delegate?

How many of these languages do you speak?

I currently speak and understand English, Spanish, Cantonese and Mandarin to varying degrees.

I was forced to take French for 2 years in high school. Never even took it seriously. But because of that, after 6 weeks of private Spanish tutoring, I was able to hold hour+ long conversations with strangers while backpacking LatAm.

I've spoken Cantonese my entire life (but not truly native level). I took an entire year of college level accelerated "Mandarin for Other Chinese Language Speakers". I took it quite seriously. I'm backpacking China right now. I still can't even talk to anyone for more than a couple minutes without having to use a translator or look up words.

> there are plenty of people in Guangzhou who are perfectly bilingual

There are also plenty who move to Guangzhou and Shenzhen and can't pick up Cantonese at all. Turns out having an authoritarian government force Mandarin on you will make the Cantonese speakers bilingual rather quickly.

High-skill immigrants disproportionately found companies (whose very existence implies driving up demand in the labor market), as well as disproportionately induce new consumer demand (high-skill -> high pay -> high consumption), which again drives new companies.

Out of America (and the world's) only $1T companies: Brin and Huang are 1st gen immigrants. Nadella is a 1st gen immigrant; he brought Microsoft out of it's malaise.

(Jobs' bio father was an immigrant, as was Bezos' adoptive father)

3/5 ain't bad. (2/5 if you're a stickler)

Tesla, which at one point was a $T company and is currently pretty close, is also famously founded by an immigrant.

There are very few people in tech in America who will not have worked for an immigrant founder/CEO at some point in their career.

> roughly 1/3 of Vancouver’s condo supply was vacant.

Why do you choose to just tell lies?

I am told as of last year, if you pay for things with Visa or Mastercard credit, they give you the blue dollar rate as well.

Consumption because they're people and spend on things.

Investment because immigrants tend to make a lot of money and (some) culturally save and invest a lot.

Net exports because immigrants work and produce things.

Can you elucidate why specifically?

I'm imagining something like an GPT-4 agent that has connections to your Databricks, Snowflake, Salesforce, etc.

You ask it a question and then it iteratively queries your data and produces some an answer/some code/ a graph. Or it generates a plan that you can edit a bit and then it executes.

That sounds like it would be pretty useful copilot for me? I'm not very familiar with the intricacies of analytics work though.

There's a reason they changed Google's internal chat system to delete all messages after max 30 days. They don't want discovery happening again.

Maybe you can write a blog post or a reddit thread?

There were a bunch of people asking about it on Discord for Godot 4, and some confused reddit threads offering to pay people for a tutorial:

https://www.reddit.com/r/godot/comments/11vxeo7/can_i_please...

https://old.reddit.com/r/godot/comments/13ka6qc/will_pay_to_...

Godot currently doesn't even have good support for things like in-app purchases for iOS/Android. No one is willing to work on stuff like that for free/open-source fun.

I think the big moat is that for studios, games are so risky to make, that there's no reason to take on tech risk by possibly running into issues like this. You know Unity will work for basically any non AAA game you want to make.

Hopefully this $15m closes that gap.

You can't be serious?

Those are just frontends on top of Stable Diffusion. They won't provide what you need for a game - the results aren't animatable or rigable.

Biggest question: How much of OpenAI's IP do they get to access at Microsoft? (and perhaps take with them to whatever new startup they would obviously found after?)

How many autonomous miles has comma driven?

That's a parody account

In journalist speak, that means "credible person I know who would get fired if I said their name."

It's probably literally an exec on the OpenAI/Microsoft partnership.

The alternative to "a person familiar" is us as readers never get this information at all.

Why didn't other oil companies have a go at building something better than Standard Oil?

Both Bengio and Hinton have their names plastered over many of the seminal works in deep learning.

AlexNet, the paper that arguable started it all, came out of Hinton's lab.

https://papers.nips.cc/paper_files/paper/2012/hash/c399862d3...

I really don't think they need to build any more of a brand.

You can get something similar for free here:

https://huggingface.co/spaces/AP123/IllusionDiffusion

Yeah, because that would mean someone is subsidizing it.

What have you been trying to use agents for?

I think most of the actual value created short/medium-term with agents will be in entertainment where things going awry is OK and no action they take can really be catastrophic.

Some people are already spending a ton of time on character.ai (https://beta.character.ai). There are sims-like games like ai-town (https://www.convex.dev/ai-town). My friend and I made an open-source murder mystery novella with agents (https://gron.games).

We found that a lot of the actions they take are pretty low stakes, so the correctness issues are much less important (in ai-town, they just talk with each other, in our game, they just beat you or run away if you interrogate them poorly).

Delivery that a drone could replace are maybe 1% of the traffic MAX.

There is a new externality that they're not paying for. I don't care if other people pay for it, I want to be reimbursed for the externality it causes me.

Thanks!

> I wonder how the economics of this works - I'm guessing token markup?

We're mostly thinking about building tools now for a future in which many gamers will have the compute to locally run a sizable ~30B model. It's hard to me to see LLM gaming taking off in the very near term due to the compute costs for any model of reasonable immersiveness.

> I wrote a library for structured inference in-browser[1]

Very cool! Excited to see where this ends up. Yeah, OpenAI is a bit annoying since you can't ask for JSON or the logit distribution. But GPT3.5-turbo right now is even cheaper than running your own llama2-70b so we stuck with that for the demo since most people probably don't have sufficient compute for this.

Is it a NIMBY sentiment to not want these in cities?

I can't imagine the kind of clusterfuck these would cause in a sufficiently dense city. The sky blotted with these...