Greetings all, and happy Thursday!
I hope ye are all keeping well. If you’re in the UK, Spring was nice wasn’t it? I can’t complain though, had a lovely time away in Seville. In a move that does not play to any stereotypes whatsoever, I was delighted to discover a new potato preparation, papas aliñas. So unbelievably good. I’m becoming a big vinegar fan.
In facial recognition news, and I guess very much on the naughty step, The Guardian (and Mark Wilding, again!) have a piece on what I believe is the first case of wrongful arrest involving facial recognition in the UK. The BBC also has a piece on a shopper being wrongly identified as a shoplifter. And meanwhile, it looks like the Met are going to start rolling out operator initiated facial recognition. These are obviously very significant stories in terms of the UK’s current consultation on facial recognition.
The Biometrics Commissioner’s response to the FRT consultation is here; the EHRC’s here; and myself and Pete’s here; a Liberty comment on the consultation (in the context of expanded rollout) is here; and a Big Brother Watch comment on the rollout expansion is here.
Another week another disaster for the Dutch in terms of their risk assessment rollout. This time its news that a probation risk assessment tool had to be discontinued. This comes after the childcare benefit and student loan scandals. If you haven’t come across it, the Australian Robodebt scandal is also really significant in this area, particularly in terms of the number of people impacted (700,000 from memory).
I missed this, but apparently Anthropic was used during the US raid on Venezuela. There’s very little info in the piece unfortunately.
I’ll leave you this week with Tennessee Ernie Ford and 16 Tons (sic). You can guess what TV show I just finished watching.
Stay well, be lovely.
— — —
The Guardian, US military used Anthropic’s AI model Claude in Venezuela raid, report says
The Guardian, Facial recognition error prompts police to arrest Asian man for burglary 100 miles away first example wrongful arrest
BBC, Met using e-bikes and drones to tackle London phone theft (the expansion of surveillance drones is the interesting bit for me.)
BBC, Chester man accused of being thief after Home Bargains face scan alert
BBC, Live facial recognition to be used ahead of Everton v Man Utd
The Hill, Ring ends partnership with surveillance firm after Super Bowl ad backlash
Dutch News, Probation service used error-ridden algorithms to assess risks
The Times, The machines that will predict the criminals of the future
The Guardian, Met police using AI tools supplied by Palantir to flag officer misconduct
The De-Center, Unplugging a Nation: Iranian Digital Workarounds in the Face of Total Blackout (not ai but pretty interesting?)
The Guardian, Zuckerberg grilled in landmark social media trial over teen mental health (he came flanked with friends in Meta glasses, and the judge noted they would be held in contempt if they were recording)
Blood in the Machine, Across the US, people are dismantling and destroying Flock surveillance cameras (The Luddites advance)
The Financial Times, How tech turned against women
Progressive, The Human Cost of Unregulated AI Tools
404 Media, Leaked Email Suggests Ring Plans to Expand ‘Search Party’ Surveillance Beyond Dogs (first they came for the dogs…)
The Guardian, Google puts users at risk by downplaying health disclaimers under AI Overviews
Bleepign Computer, Microsoft says bug causes Copilot to summarize confidential emails
The Finacial Times, Sequoia leads $1bn seed round for ex-Google scientist’s new AI lab (British scientist raising $1bn for new AI lab in Europe’s biggest seed round, it applies experimental (reinforcement) rather than llm learning )
Amnesty International, Global/India: AI Impact Summit failed to rein in destructive practices of governments and technology companies
The Financial Times, Synthesia’s Victor Riparbelli: forget everything you know about video (synthetic video, interesting piece)
The Guardian, China’s dancing robots: how worried should we be? (eh, quite; there’s a video)
Tech Policy Press, The Real Cost of the UK’s ‘Free AI Training for All’ is Democracy
EFF, Amazon Ring Cashes in on Techno-Authoritarianism and Mass Surveillance
The Guardian, AI is indeed coming – but there is also evidence to allay investor fears | AI (artificial intelligence)
The Hill, Mark Zuckerberg quizzed on kids’ Instagram use in social media trial
The Hill, Instagram launches new tool alerting parents about suicide, self-harm searches
EDRI, How recommender algorithms threaten election integrity
Rest of World, AI is giving tech companies power that once belonged to governments
AlgorithmWatch, The Seamless Surveillance Machine: Europe’s Biometric Border Vision