Take it as you will, but errors on OpenAI and ChatGPT API are multiplying… and yes, inevitably, when you no longer have a highly motivated team to maintain such a giant product, it can quickly become a headache…
@dfintelligence
-
Prepare to switch from OpenAI API to other LLMs
By
–
To all AI teams and individuals who have built services on #OpenAI #API, you must be ready at any moment to switch your service to another service. (Anthropic, LLama, any LLM locally on an instance with guardrails in
-
Building a serverless API infrastructure to call LLMs locally
By
–
It will happen very soon. I need to build a serverless API infrastructure as a service to call LLMs locally on the instances.
-
AI debate Space tonight at 8:30 PM
By
–
This tweet generated a lot of reactions, so tonight I propose we hold a debate and discussion Space about AI in general. Come share your point of view at 8:30 PM https://x.com/i/spaces/1nAKEaXrpqVKL …
-
Theory: Altman fired for hiding a conscious AI from OpenAI
By
–
Il y a une théorie que Sam Altman aurait été viré parce que OpenAI aurait trouver un moyen d’avoir une IA consciente / AGI et qu’il ne l’a pas dis au board.
— Defend Intelligence (Anis Ayari) (@DFintelligence) 18 novembre 2023
Cette thèse revient souvent en plus chez de gros acteurs de la tech mais venez on souffle et on redevient sérieux 2 sc.. https://t.co/MHReYiT3CzThere is a theory that Sam Altman was fired because OpenAI found a way to have a conscious AI / AGI and he did not tell the board. This thesis often comes up among big tech players but come on, let's breathe and get serious for 2 seconds..
-
How to understand LLM and AI without Machine Learning?
By
–
Hello, I have a probably stupid question but how can you be knowledgeable about LLM and AI without knowing anything about Machine Learning? Please.
-
Explanation of leads on answer selection during training
By
–
The paper explains at the end some leads, such as answer selection during the labeling/training phase, but more research needs to be done to understand the phenomenon. Interesting. The source: https://arxiv.org/pdf/2307.11760.pdf
… -
LLMs improved by fake emotions according to Peking University
By
–
Does appealing to the (nonexistent) emotions of LLMs make them better? The answer from Peking University seems to be "yes". In this paper they demonstrate that an LLM gives better results by using terms like "this is very important for me".
-
AI Debate: Regulation, Response to Astier, One Piece TCG Card Opening
By
–
Meet on Twitch! We will discuss the big current debate in AI: should we regulate AI or not? We will talk about Astier's take to which I just responded. Preceded by a small opening of a One Piece TCG card that I just received.