Case study with the U of M knocked researchers off their feet with how much time they were able to save leveraging https://t.co/LC6BYB1qYU sign up today, it's free! Video labeling that took 10 mins, down to 20 seconds, yea that's a real metric! https://t.co/h2YccCv9aE

"Some of the most exciting developments in AI will come not from NLP alone but from integration with other technologies that help it better understand the world around us." -@_silverhare

Going back to the need for specificity in technology, @_silverhare gives the example of how important it is to have precise searching for something in healthcare - lives are at risk. Government docs will require very specific tools as well.

Today, Generative AI takes a big step and comes to user interface design! @helnzhou and I are excited to announce @Galileo_AI : the first AI product that uses natural language to generate UI designs. It lets you design beyond imagination. Early access: https://t.co/4KqV1csQ6c

This looks amazing! Congrats to the @genintelligent team for the launch and looking forward to the research coming out from these efforts.

https://t.co/w74jCg170z

AI is helping us understand the human genome, break down landfill waste, and eliminate hate speech online. The applications of AI keep getting more varied and expansive each year. How is your company using #AI?

The #LaMDA chatbot is an impressive engineering feat that requires Google-scale infra to accomplish. But the #AI community agrees it’s not "sentient". The interesting conversation is now around AI capabilities, the Turning Test, and AI hype. Do you agree?

The human language is incredibly complex and words can have so many meanings. How do YOU train ML models to handle subjectivity when a word can be taken in multiple ways? We have a few thoughts: https://t.co/xHML6IE22G #machinelearning #datalabeling

The world is going to move faster and faster from here on. People and institutions who embrace this will be richly rewarded. People and institutions who do not or cannot are going to be challenged. Adaptability and speed have gone from valuable to critical.

Large language models can be toxic or harmful to people, and some AI researchers are trying to change that

last night a neighbor i don't know well shouted to us from a distance, "saw your company was in the news again! what's up with all the sexual harassment?" like that's a normal question with a normal answer. but it got me thinking. and this might come out kinda messy. (cont'd)

Something I wish someone had told me early on... Stay Out of the Wind and Away From the Back by @kennethn

Great designers are strong at "product thinking." This is a key aspect of many design interviews, as well as many PM or VC ones. But what exactly is product thinking? And how does one get good at it? Thread below 👇

Honestly, just a fascinating read. https://t.co/o9CFqhjqL2 via @stratechery

Glad to see this newsletter back up and running. Welcome back @kennethn

https://t.co/E4aDpsPhWd The problem is the temptation to start policing everyone else because you believe that you are accountable for their failures. It is the Post that is responsible for what it publishes, and it is a mistake for Twitter to take responsibility for that.

https://t.co/XTGGbyKROD very powerful and very "obvious" use of generative adversarial networks to save a ton of bandwidth in conference calls!

Datasaur & Diffgram - Use Datasaur as Interface [Deep Learning] Demo Tut... https://t.co/kUQSlRjrxw #DataScience #Data #DataAnalytics #MachineLearning #Annotation #datasaur (2/3 on New Integrations)

Today, I'm happy to finally reveal our latest round of funding, with phenomenal backers @Initialized, @ycombinator and Chairman/CTO of OpenAI Greg Brockman. Many thanks to @ron_miller for sharing our story!