Meta AI researchers show how current language models differ from the human brain & highlight the role of long-range & hierarchical predictions. Read the open accessed article in Nature ➡️ go.nature.com/3ZfH7eK

Is this disinformation? 🤭 reddit.com/r/midjourney/c…

The suggestion that the shirt in the middle is residual of Ukraine as part of a “charity” prompt is fascinating. reddit.com/r/midjourney/c…

Finally, words in synthetic images. (Readies popcorn for the next generation deluge of misinfo and memes) reddit.com/r/StableDiffus…

"His vivacious intellect and charismatic personality lit up our seminars, conversations, teaching, and research." Vicki Nash and Bill Dutton remember Professor Yorick Wilks who sadly passed away last week: oii.ox.ac.uk/news-events/ne…

Self-supervised learning underpins today’s cutting-edge work across natural language, computer vision & more — but it’s an intricate art with high barriers to entry. Today we're releasing the SSL Cookbook, a practical guide for navigating SSL + contributing to this space ⬇️

Not bad. reddit.com/r/datascience/…

Exciting news to share! Our team has just been funded to develop Network Canvas Studio - a cloud-based version of Network Canvas. Learn more: isgmh.northwestern.edu/news/2023/clou… (1/3)

It’s just fish in a barrel evaluating the biases in these models. But it’s pretty striking. reddit.com/r/midjourney/c…

Lazy and provocative. Could have taken brief more seriously: Vogue Just Used AI-Generated Imagery in a Nightmare Fuel Cover Shoot futurism.com/vogue-ai-image…

Tears of the Kingdom reviews set a new standard in hyperbole.

The proper likeness continues: Apple speech-to-text now can include your own voice: apple.com/newsroom/2023/…

Actual working artist with actual programmer reveal that AI generated art is not a point-and-click affair but difficult, fragile, and subtle if you want something interesting and representative of the art. reddit.com/r/StableDiffus…

Next, 'The Poverty of Ethical AI', presents a case study based on fieldwork conducted at Sama, a data annotation company with headquarters in San Francisco and outsourced centres in Kenya and Uganda and reveals the human costs of purportedly ‘ethical AI’ link.springer.com/article/10.100…

Although I nearly never post on this platform, I wanted to share a link to a nice interview about my latest book, "From Social Science to Data Science", a monograph about using Python to interpret data. researchmethodscommunity.sagepub.com/blog/data-scie…

Hi! Testing LLaMa the Oxford Internet Institute 's brand new 4x H100s box for student research. It says: “The self is not a product of the brain but a product of the autoreferent. The brain is a product of the Euclidean space that the autoreferent uses to simulate being outside of itself.”

Finally packing up. Find me at blurky.bsky.social. xox