To be fair, a bubble is more of an economic thing and not necessarily tied to product/service features.

LLMs clearly have utility, but is it enough to turn them into a profitable business line?

First time I heard about Hola. Read an article about them. They claimed that:

Hola CEO, Avi Cohen, said: "Hola offers a valuable free service to its users, and our community agrees, consistently rating it higher than any other VPN service. The health of the Internet and the safety of its users are top priorities for Hola."

Avi Cohen is liar and scoundrel.

But Tor is explicitly designed for this sort of thing and does not have a paid access system.

Nowhere on Big Mama VPN's landing page does it say that by using their service you open your network to access by randoms.

Big Mama is made up of two parts: There’s the free VPN app, which is available on the Google Play store for Android devices and has been downloaded more than 1 million times. Then there’s the Big Mama Proxy Network, which allows people (among other options) to buy shared access to “real” 4G and home Wi-Fi IP addresses for as little as 40 cents for 24 hours.

I am surprised this is legal. I have zero knowledge of the nuances of law, but there is no situation where allowing a completely random 3rd party to use your IP/network on an on-demand basis is legitimate. This is not a "grey area" (in the real sense, not in the judicial sense).

Yeah, this is not even really targeting clickbait, more like putting restrictions on openly malicious content.

Just goes to show that russians have had (until recently) relatively easy access to independent information within a few clicks on their smartphones.

I will also note that many reliable news organizations (BBC, DW) started their russian language YT news programs as far back as 2010. This is also true for well regarded local independent news (TV Dozdh).

Haha. I believe the AMD Instinct / Nvidia Datacentre GPUs aren't that great for gaming.

Assuming a large decline in demand for AI compute, what would be the use cases for renting out older AI compute hardware on the cloud? Where would the demand come from? Prices would also go down with a decrease in demand.

Archive

I wonder what the process was for choosing specifically 10%. Why not 8.7%? Or 13.9%? Surely an efficiency drive would have some sort of structured/analytical approach to it?

I use the “udm=14” method to remove the AI/Spam results at the beginning of search result. While it doesn't address the overall decline in search quality, I find it is a much better experience.

Given a sufficient amount of text, the method is said to be 99.9 percent effective.

If that's really the case, they should release some benchmarks. I am skeptical. Promising the world is a key component of their "business model".

I am increasingly starting to believe that all these rumors and "hush hush" PR initiatives about "reasoning AI" is an attempt to keep the hype going (and VC investments) till the vesting period for their stock closes out.

I wouldn't be surprised if all these "AI" companies have come to a point where they're basically at the limits of LLM capabilities (due to problems with its fundamental architecture) while not being able to solve its core drawbacks (hallucinations, ridiculously high capex and opex cost).