I read it and confirm it's garbage, a computer + a camera… Anyone can do the same demo with GPT's API, capturing 1 frame every 5 seconds and adapting in real-time.
@dfintelligence
-
Critique: AI Demo Judged Disappointing, Easily Reproducible with GPT API
By
–
I read it and I confirm it's very poor, a computer + a camera…
Anyone can do the same demo with the GPT API that captures 1 frame every 5 seconds and adapts in real-time. -
Critique of scripted AI demos and unfulfilled promises
By
–
Qu’est ce qui aurait empêché de faire la démo en live ?
— Defend Intelligence (Anis Ayari) (@DFintelligence) 15 mai 2024
Je commence à en avoir plus que marre des démos d’IA faké et bien scenarisé, des annonces de produit et de fonctionnalités qui verront le jour que dans des mois (voir jamais… on connaît Google maintenant … 🙂 )
Le but ? https://t.co/GbZbxbS2JkWhat would have prevented doing the demo live? I'm starting to get more than fed up with fake and well-scripted AI demos, product and feature announcements that will only see the light of day in months (or never… we know Google now … 🙂 ) The goal?
-
Google I/O Summary: Gemini repetitive, only Gemini 1.5 interesting
By
–
If you want a summary of Google I/O: – It's Gemini I/O, all they talk about is that. With demos and features that are repetitive and 90% of which will be available in several months for some (i.e., maybe never, we know Google). – The only interesting thing is Gemini 1.5
-
Don’t trust demos, wait for GPT-4o, price in question
By
–
Never trust demo. Personally, I'm waiting to test GPT-4o. We know the "available in a few weeks". Calm down. We know we will get to a real-time multimodal AI. The real question is how and at what price.
-
SQL, game changer for generating complex queries
By
–
exactement ! et le game changer c'est le sql. Je dump le model de ma bdd , et je lui dis fait moi une query pour "…." qui parfois requiert des dizaines de colones avec des filtres, jointures etc… et il te pond une requête magnifique. pic.twitter.com/R8T7kB6CZD
— Defend Intelligence (Anis Ayari) (@DFintelligence) 13 mai 2024Exactly! And the game changer is SQL. I dump my database model, and I tell it to make me a query for '….' which sometimes requires dozens of columns with filters, joins, etc… and it produces a magnificent query.
-
Prefers ChatGPT in console rather than Copilot in IDE
By
–
I use PyCharm, VSCode I only use for C# and I've never gotten into the habit of Copilot, it stresses me out too much to have something that spits into my IDE. I much prefer having everything written in the ChatGPT console and just copy-pasting what interests me.
-
Correcting ChatGPT: 50% of development work
By
–
50% of my dev work now consists of 'just' correcting ChatGPT, pointing out its errors, and improving the code it proposes. Coding has become so comfortable. What a wonderful time to be replaced in code-spitting tasks.
-
LLMaM: Predict your target variable with an LLM
By
–
Welcome to the era of LLMaM! (LLM as a Model) Forget about models like Random Forest trained on data, just ask an LLM to predict your target variable! In 2018, we were promised autoML with heavy feature engineering and grid search in all
-
Received Rabbit R1, asking for opinions and suggestions
By
–
Guys, I just received my Rabbit R1!! Any thoughts? What should I try???? So excited.