Might not be as “good” as #GPT3, but it is truly #opensource and you can download the full model to play around with. eleuther.ai/projects/gpt-n…
#OpenAI was smart (or just cunning) in promoting their brand and GPT. #Google is still dominating actual #AI usage. thenextweb.com/neural/2021/01…
4.5 billion words a day, when it scales up to 4.5 trillion words a day, how can you tell if most content you read are synthetic? (Sound better than fake,...
Do you still trust #Google? Or just too lazy to switch to Bing or duckduckgo? How about switching to Brave or Microsoft Edge too? bloomberg.com/news/articles/…
Waiting for EleutherAI to come up with a truly open source competitor to #GPT3. You can’t stop AI development. Not releasing GPT3 is a big mistake. eleuther.ai
Will #Google release the trillion-parameter model for download? Or will it keep it guarded like #GPT3 and plan to make a lot of money from it? #AI venturebeat.com/2021/01/12/goo…
RT @TrungTPhan: When a startup claims it has “AI” or “ML” and then you take a look at the actual tech pic.twitter.com/2acvwXiJpy
Assuming that #AI will one day be unbiased and ethical, what will we do with humans not strictly abiding to it? venturebeat.com/2020/12/10/tim…
Data & #AI Landscape 2020. Useful chart to keep. mattturck.com/wp-content/upl…
Photos edited with #AI-powered Neural Filter by Adobe are socially acceptable Professional #Deepfakes? newatlas.com/photography/ph…
“Less than one”-shot (LO-shot) learning. Next gen #AI will need less data and processing power. technologyreview.com/2020/10/16/101…
How about starting each video conferencing session with a #FaceID-like setup process to get a simple high-res 3D mapping of your face? Maybe it can handle the whole call with...
Why are we so scared of fake random text? Because they are not clearly labelled as fiction on bookshelves? #AI hbr.org/podcast/2020/1…
Interesting way to split up #AI personalization models training load. AI 2.0 will focus on optimisation? venturebeat.com/2020/09/29/lin…
#Apple is pushing ahead with personalised machine learning. Creating models about you on your own device. Ready for privacy sensitive #AI computing to come. cultofmac.com/723977/apple-s…
Is the #AI fight just between #Google and #Facebook? venturebeat.com/2020/01/30/ope…
Not saying that #Google scientists are not smart enough, but using #AI for 10+ days of weather prediction is much harder than winning with #AlphaGo. theverge.com/2020/1/14/2106…
“Retirement” or just “Redundant”? #AI to replace more and more soft-skill roles? gizmodo.com/go-champion-re…
Fat people keep getting fatter, really? #AI #MachineLearning #Blockchain #GPT2 medium.com/@eplt/fat-peop…
When will quantum computing start doing something useful? Like cracking #crypto keys? Another 20 years? #quantumsupremacy technologyreview.com/s/614608/googl…
#Chatbots were supposed to take over most customer service jobs and be the first killer-app of #AI. Didn’t happen. onezero.medium.com/slack-and-the-…
It’s time to play around with GANs, Generative Adversarial Networks. #AI adweek.com/digital/patent…
Datasets to play around with. #MachineLearning #AI developer.ibm.com/exchanges/data/
When will #Netflix come up with a #Bandersnatch with your own choice of actors, all done by #AI face swapping? uncrate.com/video/jim-carr…
Great use of image recognition. But won’t stop people debating the actual number. #OpenCV graphics.reuters.com/HONGKONG-EXTRA…
Can this be considered as #DeepFake? You are not looking at the camera, but the software will change the image to look like you are. techcrunch.com/2019/07/03/app…
It’s a very big assumption that we can understand #AI. Just as world’s top Go player cannot understand AlphaGo. venturebeat.com/2019/06/14/goo…
Learn a bit more about Generative Adversarial Networks (GANs). #AI machinelearningmastery.com/impressive-app…
Human staff will always be needed to help machines to replace other human staff more effectively, they can’t do it alone yet. #AI bbc.com/news/technolog…
#WWDC2019 Wishlist: How about an #AI #CoreML anti-#deepfake technology? Auto detect and alert potentially fake or manipulated images on device?