Are all taxis yellow? A story of biases
I had the idea to use Stable Diffusion XL to create images and illustrate articles I’m about to write.
To be honest, I was not obtaining the result expected and was failing prompts after prompts.
Prompting is probably a form of art but it’s also complicated to identify which parts are assumed and covered (and how) by the model and which ones need to be detailed.
Many voices have begun to say that the gap between humans and artificial intelligence systems is shrinking quickly.
Let’s confront the state of art and challenge this statement using simple experiments.
Nenuphars, gas consumption and crypto stories? Coding is not dead. Yet.
Developers will be replaced massively by AI as coding becomes a useless task, starting today.
This idea and those statements started to grow after a few influencers experienced asking a chatbot to code a Tetris and deploying it and “it was working”.
They decided to take a big shortcut and concluded that “no-code” will be the new reality soon and the capacities of GPT and Large Language Models would make coders obsolete in the very near future.
It’s a simplistic opinion based on one single non representative use case used as a universal proof.
It does overestimate the capacities of the models we use.
Large language models are not truly performant polyglots: how to escape the english centric trap?
Many people claim that large language models and particularly ChatGPT can speak and answer in almost all languages like humans.
Those marketing messages take a big shortcut by advertising those tools as efficient polyglots.
The progress in natural language processing is real but some problems and limits do remain on the table.
Are those tools equally efficient in all languages like some of us seem to think?
Isn't English better represented and related prompts receive better outputs?
How are those models trained and how important is the choice and volume of the initial corpus? How do editors fight against hallucinations and cultural biases?
Nenuphars and gas consumption : let's ask Gemini !
In a previous article from October 2023 I was trying to test Llama and GPT's logic by trying to solve simple math problems.
https://www.kindrobot.org/article/nenuphars-gas-consumption-and-crypto-stories-coding-is-not-dead-yet/
And the results were not this impressive.
As the technology evolves and improves and as editors claim "giant" progress loudly on a regular if not daily basis, let's try to challenge one of them: Google Gemini.
Gemini, the large language model developed by Google was launched last December. A big roadshow.