Microsoft researchers published a very informative paper on their pretty smart way to let GenAI do 'bad' things (i.e. 'jailbreaking'). They actually set two aspects of the fundamental operation of these models against each other.
Tag: GPT
Memorisation: the deep problem of Midjourney, ChatGPT, and friends
If we ask GPT to get us "that poem that compares the loved one to a summer's day" we want it to produce the actual Shakespeare Sonnet 18, not some confabulation. And it does. It has memorised this part of the training data. This is both sought-after and problematic and provides a fundamental limit for the reliability of these models.
State of the Art Gemini, GPT and friends take a shot at learning
Google’s Gemini has arrived. Google has produced videos, a blog, a technical background paper, and more. According to Google: "Gemini surpasses state-of-the-art performance on a range of benchmarks including text and coding." But hidden in the grand words lies another generally overlooked aspect of Large Language Models which is important to understand. And when we use that aspect to try to trip up GPT, we see something peculiar. Shenanigans, shenanigans.
The hidden meaning of the errors of ChatGPT (and friends)
We should stop labelling the wrong results of ChatGPT and friends (the 'hallucinations') as 'errors'. Even Sam Altman — CEO of OpenAI — agrees, they are more 'features' than 'bugs' he has said. But why is that? And why should we not call them errors?
The Truth about ChatGPT and Friends — understand what it really does and what that means
On 10 October I gave an (enthusiastically received) explainer talk at the EABPM Conference Europe 2023, making clear what ChatGPT and friends actually do — addressing the technology in a non-technical but correct way — and what that means. That presentation fills the gap between the tech and the results. At the end you will understand what these models really do in a practical sense (so not the technical how) when they handle language, see not only how impressive they are, but also how the errors come to be (with a practical example), and what that means what we may expect from this technology in the future.
R&A IT Strategy & Architecture 