Greetings all, and I hope ye are keeping well. I rather over optimistically put on shorts this morning after sitting out the back in the sun over coffee. Really regretting that decision now.
So, I was supposed to be in Zambia at Rightscon this week, but Zambia cancelled the conference at the last minute. Apparently due to Chinese influence. WIRED have the story.
I also must apologise for my very rushed comment on Carlo and Thompson last week. Ouch. The typos. The repetition. Thats what you get for trying to rush something off before a long day of meetings. Lesson learned. It is an interesting case, but I think mainly for the fact that (a) it will be used to legitimise future LFR operations, but (b) it did not really address the key issues relating to whether LFR can be considered ‘necessary in a democratic society’ (and if so what uses) and how watchlists are constructed. So much – really the important stuff – is left unaddressed.
Speaking of LFR, i’m not sure how i came across this BBC video, but its a great insight into the potency of the surveillance state in China. And its from 2017. Can you imagine everything that has happened since? It’s really relevant in light of the new Policing AI initiative here, primarily (i think) because it demonstrates (or at least hints at the fact) that surveillance cameras and facial recognition are just foundation technologies. It is the AI-assisted analysis (profiling, monitoring, etc) that they enable that is really game changing/game changingly intrusive. On that note, the Register has a story on a new tool to send data on shoplifting straight to the police, for use with retrospective facial recognition. The Guardian also has a piece, with comment from the Biometrics Commissioners, on how regulation is lagging far behind FRT use.
And speaking of AI-related surveillance, LIghthouse Reports have a new investigation into an AI system deployed by the Kenyan government to set healthcare costs. Worth a read. And the flawed use of proxies bears striking similarities to HRW’s previous reports on social assistance in Jordan.
The deadline for CAIDP’s Autumn AI Policy Clinics is approaching soon, 15 May. Info at the link.
The Hill has a story on major AI firms signing a deal with the Pentagon to deploy their systems for ‘lawful operational use’. Given the US’s current near total disregard for international law, it’s hard to see how this would pass any human rights due diligence test.
I’ll leave you this week with Irish Goodbye by Kneecap (thats a link to the short movie they made around the song, which is quite moving, the song itself featuring a great spot by Kae Tempest is here). Solidarity and love to everyone experiencing suicidal ideation, or affected by suicide.
Stay well, be lovely.
Lighthouse Reports/Africa Uncensored/The Guardian,Hiding Behind AI: How SHA Was Used to Load Health System Costs Onto Poorest
The Guardian, AI facial recognition oversight lagging far behind technology, watchdogs warn
The HIll, Pentagon reaches deal with leading AI companies for classified work
Byline Times, Decoding the Technofascist Tendency in Palantir’s Manifesto
WIRED, Palantir Employees Are Starting to Wonder if They’re the Bad Guys
Tech Crunch, Surveillance vendors caught abusing access to telcos to track people’s phone locations, researchers say
Tech Policy Press, Orbán’s Hungary Defeat Shows Disinformation is Not a Political Magic Trick
The Guardian, What is Mythos AI and why could it be a threat to global cybersecurity? (included after all too brief mention the other week)
The Hill, Anthropic becomes impossible for White House to ignore (more on Mythos)
The Guardian, AI-powered robot beats elite table tennis players (if their table tennis players are this good, what are the military robots like?)
RUSI, Mobilisation and Training for War: Preparing to Break Glass
NPR, Ingenious? Orwellian? Or both? Supreme Court considers constitutionality of ‘geofence’ warrants
The Guardian, Met investigates hundreds of officers after using Palantir AI tool | Metropolitan police
The HIll, Google rolls out AI tool to predict flash floods up to 24 hours ahead
UNICRI, Decoding Transparency – How to Foster Public Trust in Responsible AI Innovation in Law Enforcement
EDRi, Greece’s AI Smart Policing system ruled unlawful after €4 million public spending\
Rest of World News, I got stood up by an AI agent, and tracked down its human owner in China
Rest of World News, Five times AI hallucinations embarrassed governments
BBC, Appleby Horse Fair police operation to use face recognition
The Register, Met police trials new tech to counter shoplifting scourge
BBC, Three arrests in Slough after police live facial recognition use
BBC, Shops asked to share more CCTV to help police tackle shoplifting
The Guardian, Scanned, tackled, arrested: how live facial recognition was piloted on the streets of Croydon
The Guardian, How does live facial recognition work and how many UK police forces use it?
WIRED, Exclusive: Metalenz Has Figured Out a Way to Make Face ID Invisible
Financial Times, How bots could help revive democracy (bots have been great so far…)
The Guardian, Pentagon plans to make US military ‘AI-first fighting force’ by pairing with companies
Financial Times, Paranoid parenting in the age of AI
WIRED, Disneyland Now Uses Face Recognition on Visitors
The Register, Five Eyes spook shops warn agentic is too wonky for rapid rollout (who writes these headlines?)
Financial Times, Restaurants lean on AI to cut waste and reduce costs
New York Times, A.I. Is a National Security Risk. We Aren’t Doing Nearly Enough.
The Guardian, ‘Think before sharing,’ Giorgia Meloni says as AI-made lingerie image of her goes viral
The Register, India orders infosec red alert in case Mythos sparks crime spree
The Washington Post, After 5 days on a D.C. bridge, a man ends his protest and comes down (there’s an AI link, honest)
New York Times, Five Ways A.I. Search Beats an Old-School Google Search