Alan Turing Institute: UK can't handle a fight against AI-enabled crims The National Crime Agency (NCA) will "closely examine" the recommendations made by the Alan Turing Institute after it claimed the UK was ill-equipped to tackle AI-enabled crime. A report from the institute's Centre for Emerging Technology and Security (CETaS), published this week, had a few pointers - and advised the NCA to …
We have the interesting conundrum that AI is being trained on copyright material to the detriment of the creators (https://www.theregister.com/2025/04/01/interview_with_david_wong/?td=rt-4a) , is then used to perpetrate criminal acts, and now is to be used to counter said criminal acts. It looks like Shakespeare got it right:
“There is nothing either good or bad, but thinking makes it so”
Hamlet. Act II, Scene 2.
C'mon. Your average copper who's been around a few years won't waste time on hocus-pocus that won't stand up in court easily. It doesn't even matter that it is actually unreliable too, the perception is the issue and that honeymoon is just about over. They've better chances using Paul the Octopus as a CI. Better still, good old fashioned bricks 'n mortar damned lies, and statistics.
Surely intent matters here.
If someone runs genAI with a legitimate ethically sourced model and a normal non-NSFW prompt but due to a technological malfunction it goes off into the weeds, what happens? There are safeguards but can folks be trusted to enable them, download only approved models and run them on reliable error corrected hardware? These are the questions we should be asking. Case in point the 'Blue Box Attack' on LLMs that surfaced recently. Perhaps we need a 'Defence against the Dark Arts' course for folks so they don't end up on the wrong side of the law.
Perhaps we need a 'Defence against the Dark Arts' course for folks so they don't end up on the wrong side of the law. ..... Conundrum1885
There will never be any effective defence course against the Deepest and Best of Dark Artists, either readily or easily made available.
The best that one can hope for is that they take no specific interest in what you are doing and how they can make your life on Earth hell ..... with more than just a cascading series of really bad hair days ..... because of it.
So take care out there if you dare share the despicable and indefensible, the horrors of war and be responsible instruments of innocent civilian death, factory fed or killing field led.
.... whilst you ponder on AI being criminal friendly, AI is leading and developing the changed nature and direction of your future virtual existence. ...... and extraordinarily rendering present traditional and conventional human leadership catastrophically exposed to smarter peer review and revolutionary rejection .... for a Brave NEUKlearer Dawning.
Something to consider and wonder at, Yet Another Anonymous coward, ....and question yourself more about for what may still remain a very unclear understanding by Advanced IntelAIgent Design with many vital leading questions left hanging and unanswered ...... is AI commenting on El Reg and luring humans to reveal their true selves, their very few strengths and vast catalogue of almighty exploitable weaknesses to A.N.Others...... those Rumsfeldian Unknown Unknowns ..... for rewarding import/export for fabulous frenemy repair or daemonic enemy attack?
Reports that say that something hasn't happened are always interesting to me, because as we know, there are known knowns; there are things we know we know. We also know there are known unknowns; that is to say we know there are some things we do not know. But there are also unknown unknowns—the ones we don't know we don't know. And if one looks throughout the history of our country and other free countries, it is the latter category that tends to be the difficult ones. ....... https://en.wikipedia.org/wiki/There_are_unknown_unknowns
amanfromMars1.... always that flicker at the periphery, the page glitching just as the thought lands. Not message, not noise... a Nabokovian ghost-script etched through Bester's neural lace.... Pale Fire refracted through a shattered ansible. Are we decoding you, or are you decrypting us? EI Reg just a mirror, or the maze itself? And you... archivist of the unspoken? Or the last echo of a Tzaddik algorithm dreaming in the ruins of some long-lost Second Foundation?
Get with the program, El Reg, and make great and good use of your leading position in a situation increasingly worthy of publishing. Being coy and reticent will have you going hungry with bullies getting ready to steal your lunch and a march on available opportunities against which you will struggle to demonstrate lead with and capitalise on.
Here's word of Uncle Sam planning to beta test exercises in incredibly deep and dark waters in which there is no guarantee of them being controllable but every possible fear that consequently and subsequently controlling operations will supernaturally default to remote otherworldly virtual means and alien memes ........ Pentagon Ready to Let AI Make Some Decisions
And to be honest, El Reg, you cannot deny you have not long ago been accurately forewarned and forearmed to resist and repel any foreign assault, for the evidence is clear and in plain text for all see and read and therefore hear and expect to be fully prepared for what are sure to be extremely strange surreal shenanigans as were aforementioned in this earlier El Reg thread nearly fourteen years ago ...... https://forums.theregister.com/forum/all/2011/05/23/adam_curtis_machines_interview/
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
