shisisms

Side point and short observation, but it's now happened to me 3 times in a week.

I've found that where previously i'd leaned on 3rd party APIs to do a range of tasks that previously would have been a pain to either code, hack together or just do - I just end up using an AI sidekick to help me handle the integration.

Obvious pitfalls: nowhere near as comprehensive or thorough as the likely service i'm trying to replicate. And no doubt this serves an initial purpose only.

That said, for services that were in the $100 p/mth range, i've rapidly found myself re-creating all the functionality and crucially in a much much more custom/relevant-to-my-app way that means I just don't need them anymore.

There's an entire swathe of service that just won't have the relevance they previously had because they can be integrated with such ease.

Sounds like hyperbole, but it's happening more and more often and was curious what others were doing!

Following the long discussion as to whether GPT4 in ChatGPT is degraded. What seems to be occurring is that when you reached your limit previously you'd get a large notification saying that model has been switched.

They don't do that anymore and instead switch the model without notifying you or updating the default model notification at the start of the conversation.

So:

(a) it is degraded as it switches to 3.5-turbo

(b) it's evident in the speed of the inferences

(c) the quality drops as the context window is halved

(d) as you're notified of the change, it appears as if all is normal (as gpt4 is definitely faster than before), but in reality the instructions are followed less impressively and the context is lost faster

(e) this probably also accounts for the difference in experience across the comments as some people won't have hit the limit and been switched and some people will have switched.

No judgement on them doing this, but does explain the difference in experience.

One of the immediate concerns was the likelihood that someone would have leveraged LLMs to create a fully blown news site. But it seems like no-one has, aside from the proliferation of low level content, has anyone coherently put this into a single purely AI driven media outlet yet?

I’m curious if anyone’s tried both and if there’s any significant difference. OpenAi’s GPT API is pretty inconsistent, but wondering they’re both effectively driving from the same place.