The Brain and Artificial Intelligence

Author

Bobby Brazen
November 14, 2022

Here's your daily briefing:

  • One of the coolest things about artificial intelligence is the interplay between our understanding of artificial and human intelligence. The more we learn about the brain, the better we're able to understand AI (and vice versa). It's a mutually beneficial dance between fields that is amazing to watch. It's fair to say we're in a golden era of sorts, with so many fields converging into and feeding off each other, such as cognitive science, computer science, neuroscience, engineering, imaging, etc). And on that note:

A New Brain Model Could Pave the Way for Conscious AI

A new model of the human brain. A new study presents a new neurocomputational model of the human brain, which might shed light on how the brain develops complex cognitive skills and advance neural artificial intelligence research. An international team of scientists from the Institut Pasteur and S

https://scitechdaily.com/a-new-brain-model-could-pave-the-way-for-conscious-ai/

Anchoring future computational models in biological and social realities will not only continue to shed light on the core mechanisms underlying cognition, the team believes, but will also help provide a unique bridge to artificial intelligence toward the only known system with advanced social consciousness: the human brain.

  • Here's yet another amazing FREE resource for learning AI, from Microsoft. Check out the announcement video:

  • This is epic and we're dying to know the prompt:

  • Missed this a few days ago but this is seriously cool and worth sharing:

  • We touched on GPT-4 rumors last week. Here's a Medium post from Alberto Romero that draws them out in a little more detail:

GPT-4 Rumors From Silicon Valley

People are saying things...

https://thealgorithmicbridge.substack.com/p/gpt-4-rumors-from-silicon-valley

But, if GPT-4 is to GPT-3 what GPT-3 was to GPT-2—which isn't at all unreasonable given that OpenAI took its time with this one—this is big news.

Think about it: as models improve, we need larger leaps in performance to feel a similar degree of excitement. Thus, if we assume Scoble’s source relies mainly on perception (in contrast to rigorous scientific assessment) those claims may imply a significantly larger leap than GPT-2 → GPT-3.

In case you missed the following tweet from Sam Altman... imagine if the Turing test-passing "technological terror" is GPT-3 and "the force" being referenced here is GPT-4. If so, seems safe to assume Altman finds the difference between the two to be quite large.

  • Buildspace.so, which just raised 10 million from A16z, is hosting another free workshop on November 27th, this one for building your own writing assistant:

  • In case you ever wondered what Moe from The Simpsons would look like as a human:

And, along the same lines, this illustrator used AI to imagine what certain (human) historical figures would look like if alive today:

The illustrator uses artificial intelligence to imagine what historical icons would look like today - Karly

The photographer used artificial intelligence (AI) to imagine what historical figures like Al Capone, Benjamin Franklin and Sharon Tate would look like if

https://karlywegner.com/cbmibwh0dhbzoi8vcgv0yxbpegvslmnvbs8ymdiylzexlze0l3bob3rvz3jhcghlci11c2vzlwfplxrvlwltywdpbmutd2hhdc1oaxn0b3jpy2fslwljb25zlxdvdwxklwxvb2stbglrzs10b2rhes_saqaoc5/

Amelia Earhart:

Benjamin Franklin:

Sharon Tate:

"pencil drawing, 1970s punk rocker with a mohawk, sitting in front of a classroom of children reading them a book, black and white"