Join us next week for an Insights Session as part of our Community for Racial & Ethnic Equity in AI, focused on standards for data diversity and lessons from AI fairness assessments. Register here:

4 ways OpenAI appropriates labor & externalizes costs of ChatGPT: – authors of training data not compensated – workers in Kenya paid $1–$2/hr – those who identify failures not acknowledged/compensated – civil society, academia, regulators left scrambling to adapt/mitigate risks.

As much as Ted Chiang's fiction is fantastic, it's really his non-fiction that I could read forever. His two paragraphs here about using ChatGPT as a writing tool are the most thoughtful comments I've seen about its utility and the practice of writing. https://t.co/AVn3wbFH5V

1/7 You may have seen that we were due to begin negotiating with mgmt. Two days before our first meeting, ignoring 86% of us voting for a union and a say in our working lives, mgmt said they’re not interested in discussing anything beyond the legal minimum with us, and cancelled.

🥁It’s the moment we’ve all been waiting for… the RSA ballot results are in! With a whopping 86% of the vote, and 85% turnout – the RSA have voted IN FAVOUR of recognition! You can read all about it in the latest from @TomWall and @theGuardian here…. https://t.co/v9JyDgF9k9

1/4: **Published today!**, @Ipsos UK, @ODIHQ, @Ldn_ICHP public dialogue exploring how people feel decisions should be made about access to their health data for AI research.

1. Yesterday the DOJ won a major antitrust case. A judge blocked the merger of two publishing giants — Penguin Random House and Simon & Schuster. The ruling is great news for authors and readers. It’s also a big deal because of what it says about where antitrust is headed.

Sorry we were quiet for a bit, but we're back! With a quick request (and other updates to follow). Any lovely FRSAs who follow us on here & support our mission, we would LOVE if you could show your support for us at tomorrow's Fellowship Town Hall:

In partnership with Health Education England, the NHS AI Lab has a new report out today on developing education and training pathways to improve healthcare workers' confidence in using AI. You can read a copy here:

So despite a petition, a survey, and a majority vote showing that @theRSAorg staff would like to unionise with @iwgb, management are refusing voluntary recognition and insist we'd be better off with a staff representative group🙄Here’s why we want union recognition instead: a 🧵

Join us next Tuesday for our second Insight Session as part of our Community of Practice for Racial & Ethnic Equity in AI. The focus will be on managing data limitations in AI research for minority ethnic populations. Sign up below ⬇️ https://t.co/q8RphhjF49

🚨Important update - please RT🚨Thank you! We were blown away by the support we received in the last few days. Sadly, one of our union colleagues involved in the Observer article on Sunday was prevented from returning to work on Monday as a consequence of their involvement. 1/6

We’re the RSA union, a majority collective of staff working with @IWGBunion towards recognition. You can find out more about why we’re seeking recognition in today’s @ObserverUK by @_tomwall and here’s a short🧵on how you can support us 1/6 https://t.co/WFcVMkVlhC

https://t.co/x1PSMJTrCN

It was great to meet @EastEndTrades to discuss working together to make Tower Hamlets the best place to do business. Small business and thriving trade have always been the economic and cultural lifeblood of this Borough, and I will do all I can to promote and protect them.

🚨Today, we've published a new guide on what happens to health data & how large datasets are used. Co-created with experts & the public, it explains key terms like anonymisation & pseudonymisation. Download it, share it & use it in your work ⬇️ https://t.co/0uXXAWFxMA

Please help us amplify 📢 the STANDING Together Delphi (open until 10th October): this is our proposal for health dataset standards, to tackle issues of #AI bias and health data poverty.

Wow – onshore wind legalised. Kwarteng didn't mention it in his speech but that is potentially a big step for getting more energy supply online. https://t.co/SWXw6HkNdf

One week left to apply for legal/policy post-doc @UniofOxford @oiioxford! Work w/ me @b_mittelstadt @c_russl on an exciting project on trustworthy AI, fairness & interpretability in ML & product liability. NB we sponsor visas! https://t.co/F8DLTp9Eqc Deadline: 12:00 BST 27/09/22

Today we’ve launched an updated version of the Evidence Standards Framework. It includes evidence requirements for artificial intelligence and data-driven technologies which will help developers and innovators guide their evidence generation. Learn more: https://t.co/g3zqMyXjqr