Wrenching news: Dan Dennett has died.

He's been a great friend and incredible inspiration for me throughout my career. I will miss him enormously.

dailynous.com/2024/04/19/dan…

Again, slapping a disclaimer on a healthcare chatbot doesn’t make it responsible AI.

you know what makes one feel old, it's not becoming parent or seeing the kids grow up, it's one's parents, our teachers, and people whose works left imprints in ourselves becoming fragile and die

An arduous issue: how to create technical standards that align with human rights. A complex question with technical, legal and political dimensions internet.exchangepoint.tech/standards-huma…

'How long can Wall Street’s artificial intelligence-fuelled rally continue without clear evidence that generative AI is giving a meaningful lift to business?'
Question of pay-off from AI hangs over Big Tech earnings ft.com/content/bb1bdc… via @ft

Credit scores have been used for decades to assess consumer creditworthiness, but their scope is far greater now that they are powered by algorithms: (..) they increasingly affect whether you can buy a car, rent an apartment, or get a full-time job.
technologyreview.com/2020/12/04/101…

'Follow the (personal) data: Positioning data protection law as the cornerstone of EU’s ‘Fit for the Digital Age’ legislative package' is a chapter I wrote for the special book announced by the EDPS to celebrate its 20th Anniversary this year. Pre-print 1/ papers.ssrn.com/sol3/papers.cf…

MLR Forum be like

mireillemoret Financial Times A sure fire way to speed up #GlobalWarming

International Energy Agency: ' #electricity consumed by data centres globally will more than double by 2026: more than 1,000 terawatt hours, roughly equivalent to what Japan consumes annually'

Booming #AI has a big role

Aldo Ceccarelli mireillemoret Frank Pasquale Financial Times Aldo Ceccarelli Hallo, the unroll you asked for: threadreaderapp.com/thread/1780109… See you soon. 🤖

I wish the @ft would check the case law of the European Court of Human Rights, before suggesting that European values align with ever more investment in Large Whatever Models. See the updated fact sheet of the Court on climate change case law: echr.coe.int/documents/d/ec…

This is so spot on. I feel like almost every mainstream marketed use of generative AI is just totally brushing a fundamental part of how they (mis)function under the carpet and there is just zero plan on how to address that issue that is actually undermining almost all uses

Humans have intelligence. They can 'make sense of things they’ve never seen before'. #LLMs have memory instead of intelligence: They recall patterns but 'they don’t understand things that are different from what they have learned'François Chollet Ruth Fulterer nzz.ch/english/google…

If this is a genuine concern, then the data isn’t properly “anonymised”, and EDPB should be looking carefully at it >> ‘if “training runs” occur in the UAE and Saudi Arabia, where there are less stringent privacy protections…’ /1