which does support the idea that there is a limit to how good they can get.

I absolutely agree, im not necessarily one to say LLMs will become this incredible general intelligence level AIs. I’m really just disagreeing with people’s negative sentiment about them becoming worse / scams is not true at the moment.

I doesn’t prove it either: as I said, 2 data points aren’t enough to derive a curve

Yeah only reason I didn’t include more is because it’s a pain in the ass pulling together multiple research papers / results over the span of GPT 2, 3, 3.5, 4, 01 etc.

That’s definitely valid, but just because a tool is used for scam doesn’t inherently mean it’s a scam. I don’t call the cellphone a scam because most my calls are.

The jump from GPT-4o -> o1 (preview not full release) was a 20% cumulative knowledge jump. If that’s not an improvement in accuracy I’m not sure what is.

Compare the GPT increase from their V2 GPT4o model to their reasoning o1 preview model. The jumps from last years GPT 3.5 -> GPT 4 were also quite large. Secondly if you want to take OpenAI’s own research into account that’s in the second image.

Curious why your perspective is they’re are more of a scam when by all metrics they’ve only improved in accuracy?

with the Fediverse being a couple million people

Yes thank you as I never see this being mentioned around here. I believe the last monthly active users was 300k? While it’s not nothing, meta doesn’t give any shit about a one time acquisition of 300k users who don’t want to use their platform.

We aren’t even “worth” extinguishing right now. These decisions are being made at a level far above us. We just don’t know why yet. I’m expecting them to use ActivityPub as a tie in for Threads/Insta/FB.

Next point equally dumb: no one owns the fediverse, sure. But if enough instances say no, that means they are not welcome. Democracy and all…

If you want to talk about democracy, technically they would have the most weight as they have the most active users.

that means they are not welcome.

Also to this specifically. Not a single CEO or threads user cares.

Is it a for you based on your instance or based across all mastadon instances?

Why should Stack Overflow (part of Prosus - https://en.wikipedia.org/wiki/Prosus ) be welcomed when Meta / Threads is shunned for being a big corporate evil company?

Being shunned from the Fediverse for being a corporate (evil or not) company is an inherently bad way to think about the Fediverse.

The beauty of the Fediverse is that it’s open to everyone and free (minus hosting prices/your data) to join, if we suddenly say we’re banning every corporation, then we’re a bit of a joke of a platform as that’s the exact opposite intention. While defederation exists it’s a joke to assume every huge instance will decouple from every major corporation.

Yes it is not optimal that an application such as Threads can adopt the protocol, but generally unless they destroy the protocol itself (highly unlikely) it’ll only affect competitors in direct competition like Mastadon. Even whilst being affected the main user base should remain as it’s currently made up of like minded individuals who despise corporations, respect our data privacy etc

I’m generally pretty anti-corporation, but federating and working with corporations is a great opportunity for the Fediverse. Corpos like Twitter, reddit, and meta would provide an easily accessible path to be in the Fediverse for non experienced users. And the Fediverse in return gets more users + more content etc.

You essentially get the opportunity to keep your data out of the hands of corporations(pseudo anonymously + open source) while being supplied with content from the average users.