Thank you Kate!
Posts by Ciaran Martin
I've spoken a bunch recently about experiments with agentic AI to do genuinely useful stuff, such as "Bruce", who now replies to most of our Zendesk tickets. So, what happens when you let it loose on the HIBP API? Some pretty cool stuff: www.troyhunt.com/heres-what-a...
There is a pattern here. Both Chris Wormald and Olly Robbins appointed by this Government. Both sacked by press release when Starmer under pressure. Robbins may have approved Mandelson but it’s not credible that he acted alone w/o instruction from No10.
www.theguardian.com/uk-news/2026...
Hugh Sexey's Thing
Red for real. I was going to say this wasn’t going to finish with 22. Should have
Still 4-4 on agg but more by accident. It’s carnage
Dems projected to pick up 8-12 seats in the College of Cardinals
The best moment of the first half was when a sporting player lamped it on the volley from one penalty box to another
Has anything happened yet?
In a meritocracy, Bayern and Real would both go through and Arsenal and Sporting would be expelled from the competition for crimes against football
😂
Dedicated to @nicktheinventor.bsky.social
…then switch on HBO Max & replay the first half of Arsenal v Sporting 2/2
It’s been a highly stressful few months for everyone. War. Economic crisis. Terrifying AI breakthroughs.
At times like this, many search for the comfort of the ordinary & the predictable.
So if you’re searching for the soothing tedium of nothingness, for a world without drama or ambition…1/2
Fair
God this is dull
The other game is a banger. Mostly of comically bad goalkeeping but still
Exactly. Also I think you’re safe for as long as sporting don’t start putting it in the mixer
Pleasantly violent so far your match
Sadly not. Also lied. There is a key moment in your game. It’s labelled “start of first half”
Switched it on after 15. Clicked on “key moments”. Nothing
Switched to Bayern v Real. Clicked on key moments. A mountain of footballing chaos
Thank you James
This thoughtful thread is worth your time, on the intricacies of detangling AI hype and scaremongering – but perhaps more significantly on how the UK government *can* still do things, even if it often doesn't feel that way.
#OpenAI shots fired
“We don’t think it’s practical or appropriate to centrally decide who gets to defend themselves. Instead, we aim to enable as many legitimate defenders as possible, with access grounded in verification, trust signals, & accountability.”
openai.com/index/scalin...
While the watchword of the day is Mythos, there is a ton of other critical infosec news in today's Metacurity you should know, including
--US agencies court Anthropic AI for cyber defense despite Pentagon ban, 1/4
www.metacurity.com/us-agencies-...
here's how the FBI extracted deleted Signal messages from an iPhone:
Watch more about this, and subscribe, here: www.youtube.com/watch?v=qNUO...
10 April marks the anniversary of the Belfast/Good Friday Agreement. To mark its 25th anniversary three years ago the Society commissioned three blogs examining its legacy and present. You can read the first here:
consoc.org.uk/the-belfast-...
Bruce Schneier seems to take a similar standpoint that the issue is less Mythos (that's just great marketing), but more AI in general.
www.schneier.com/blog/archive...
Also by him & co on this topic:
www.schneier.com/blog/archive...
labs.cloudsecurityalliance.org/mythos-ciso/
Why not, their analysis is brief and easy to digest and doesn't seem to confirm the notion of a paradigma shift.
It seems to show that this is a topic to take seriously, but more because of the general trend of model capabilities and less because of Mythos.