The second AI Safety Summit, a primarily virtual event co-hosted by Britain and South Korea, opens on May 21 as hype around artificial intelligence’s potential gives way to questions over its limitations reut.rs/3UDOayj

Scoop (now free to view): Rishi Sunak’s AI Safety Institute is failing to test the safety of most leading AI models like GPT-5 before they’re released — despite heralding a “landmark” deal to check them for big security threats 👇

politico.eu/article/rishi-…

We are delighted to announce that Gaia Marcus has been appointed by Nuffield Foundation as the new Director of the Ada Lovelace Institute.

Read the full announcement⬇️
adalovelaceinstitute.org/news/gaia-marc…

Ada welcomes the Government’s plans to invest to ensure the public sector benefits from AI. However, to realise the full potential of this investment, the needs of people and society must be prioritised.

Read our full statement here: adalovelaceinstitute.org/press-release/…

#Budget2024

AI may offer opportunities to address public sector challenges, but the Government must take action to develop and deploy them responsibly, safely and ethically.🧵

The responsible integration of powerful AI foundation models into #localgov will require a process of continual learning and iteration, says Elliot Jones Ada Lovelace Institute

➡️bit.ly/3UNFzJx

SCOOP: Thousands of people using the London Underground had their movements, behavior, and body language watched by AI surveillance software designed to see if they were committing crimes or were in unsafe situations, documents obtained by WIRED reveal. wired.trib.al/GHUMy1f

The UK Government has published its response to its AI Regulation White Paper consultation.

While the Government has made some positive steps, more needs to be done to ensure that AI works for people and society.

Read our full comment⬇️
adalovelaceinstitute.org/press-release/…

British PM Rishi Sunak is dithering on AI as the US and EU forge ahead. bit.ly/3UA97ub

‘No rush to legislate’: Why is the UK so slow to regulate AI? theguardian.com/technology/202…

Don’t wait for Post Office-style scandal before regulating AI, ministers told theguardian.com/technology/202…

UK AI regulation framework emerges as government asks watchdogs to assess technology

techmonitor.ai/technology/ai-…

👩‍💻 Tech sector figures have warned that the UK government's approach to preventing the risks of AI could still leave the country 'powerless' without urgent legislative action
politicshome.com/news/article/g…

🇪🇺The EU #AIAct is nearing completion; however, EU member states still need to approve the final text.

Read our statement on why the legislation must be approved to support EU innovation and protect people and society⬇️

adalovelaceinstitute.org/press-release/…

Artificial Intelligence (AI) is developing at a rapid pace and can be found in a growing range of everyday applications. Our latest POSTnote looks at the policy implications of AI, including economic and societal benefits, potential risks and regulations: post.parliament.uk/research-brief….

NEW: Ministers will soon publish key 'tests' for when they would enact curbs on powerful AI models created by leading companies such as OpenAI and Google, as they continue to resist creating a tougher regulatory regime. W/Cristina Criddle
ft.com/content/616300…

How does AI work?
What factors are driving advances in the technology?
What is AI capable of?

Our new POSTbrief, 'Artificial intelligence: An explainer' looks at how AI is used and how it's developing, as well as perceptions and concerns. Find out more: post.parliament.uk/research-brief….

AI Act, Product Liability Directive, Platform Workers’ Rulebook reach political agreement dlvr.it/T09kPp

Medical regulators have long applied rigorous processes to new technologies that, alongside possible benefits, could present risks for people and society.

Our new paper explores lessons FDA oversight can provide for AI foundation model governance: adalovelaceinstitute.org/report/safe-be…

📢How could technology companies meaningfully involve public perspectives in AI development, and how do they understand or use public participation methods?

Our discussion paper aims to inform the conversation around ‘participatory AI’.

adalovelaceinstitute.org/report/going-p… (1/6)