Anthropic Makes 'Jailbreak' Advance To Stop AI Models Producing Harmful Results
AI startup Anthropic has demonstrated a new technique to prevent users from eliciting harmful content from its models, as leading tech groups including Microsoft and Meta race to find ways that protect against dangers posed by the cutting-edge technology. From a report: In a paper released on Monday, the San Francisco-based startup outlined a new system called "constitutional classifiers." It is a model that acts as a protective layer on top of large language models such as the one that powers Anthropic's Claude chatbot, which can monitor both inputs and outputs for harmful content.
The development by Anthropic, which is in talks to raise $2 billion at a $60 billion valuation, comes amid growing industry concern over "jailbreaking" -- attempts to manipulate AI models into generating illegal or dangerous information, such as producing instructions to build chemical weapons. Other companies are also racing to deploy measures to protect against the practice, in moves that could help them avoid regulatory scrutiny while convincing businesses to adopt AI models safely. Microsoft introduced "prompt shields" last March, while Meta introduced a prompt guard model in July last year, which researchers swiftly found ways to bypass but have since been fixed.
Read more of this story at Slashdot.
I was diagnosed with killer cervical cancer at 23 - doctors missed my subtle symptom because I was 'too young' to suffer the disease
Despite requesting a smear test from her GP, Vikki Ellis' concerns were dismissed as she was deemed too young for a screening which is only offered to over-25s.
Royal fans left in awe of six-year-old Prince Louis's 'amazing' photography skills as he takes a stunning photo of his mother Kate Middleton
Royal fans have been left delighted after it was revealed that cheeky six-year-old Prince Louis was the photographer that captured his mother, the Princess of Wales, in a poignant portrait.
Family leaves heartbreaking tribute to 'kind, thoughtful' student killed in Colchester crash
Makyle Bayley was tragically killed on Saturday morning alongside three other students from The University of Essex
Kate Middleton's 'forest bathing' technique COULD help beat cancer, say experts
The Duchess of Cambridge, 43, has long been an advocate for forest bathing, well-before her shock cancer diagnosis was revealed to the world in March last year.
Killed by ketamine in less than four years: Heartbreaking decline of coffee shop owner who went from 'happy and sociable' to a 'virtual recluse' who needed the toilet every two minutes before tragic death
Jamie Boland (pictured), founded and owned Ancoats Coffee Co, an independent Manchester roastery and speciality coffee shop.
Martin Lewis reveals clever trick to double your annual leave this year - but you'll have to be quick
The money saving expert, revealed the hack in his most recent newsletter and it couldn't come sooner for many who are yearning to get away from the gloomy UK winter.
I was left with painful, swollen arms after breast cancer surgery. But this pioneering procedure reversed it...
Lifting a hairdryer again after more than a year of breast cancer treatment, hairdresser Victoria Lock began to feel a heavy, dragging ache in her left arm.
Three stranded walkers saved just moments from death as 10-meter tide surges towards them after their tidal app mistook Silverdale, Lancashire, for Silverdale, USA
The trio got stuck in waist-deep 'fast moving' water after setting set off with their two dogs for a walk across the sands of the bay on Sunday, confident that their 'tidal app' would keep them safe
The Gigabucks Going Into Datacenter Gigawatts
The only thing that takes longer to bring online slower than a datacenter is a chip foundry, which is unfortunate for a lot of different reasons. …
The Gigabucks Going Into Datacenter Gigawatts was written by Timothy Prickett Morgan at The Next Platform.
Can the newest AirPods really detect if you're losing your hearing?
Our fact-filled, fluff-free page of travel, fitness or wellness
Altadena Girls founder, 14, who worked with Prince Harry and Meghan Markle to help LA fire victims, breaks silence after Sussexes were branded 'disaster tourists'
Avery Colvert, 14,said that support from Prince Harry and Meghan Markle's Archewell Foundation had helped turn Altadena Girls into a full-blown non profit organisation.
Two lenders cut mortgage rates ahead of Bank of England decision on Thursday
The Bank of England is expected to cut interest rates when it meets later this week, which could lead to further mortgage rate reductions.
Aristocrat who 'advised online safety bill' made crass jokes about rape and death of women on social media
An investigation into Benedict Westenra, 41, by MailOnline revealed the 8th Baron of Rossmore - who is a musician and composer - joked about rape and the death of women on social media.
Loved Miss Austen and already want more? Here are 9...
It is a truth universally acknowledged that our critics know their stuff. And here, they bring you the very best of Jane Austen available on your screens...
Charlotte Crosby reveals the name of her adorable baby girl after she confirmed the arrival of her second child with fiancé Jake Ankers
Charlotte Crosby has revealed the name of her new baby with her fiancé Jake Ankers, after sharing the first snap of her newborn girl.
Final video of talented student dancer, 21, killed in horror crash with three university pals is revealed - as heartfelt tributes pour in
The final video of a talented dancer, Eva Darold-Tchikaya, 21, who was killed in a horror crash alongside three university students has been revealed as heartfelt tributes flood in.
Kanye West fans think he is on Ozempic as he looks 50lbs slimmer at Grammys... and is asked to go 'naked' next year
'Yo is Ye in OZ?' asked one fan. Another chimed in: 'Kanye has gone Ozempic crazy like the rest of Hollywood, c'mon bro!' He was with his wife Bianca Censori who had on a sheer dress.
Backlash as Gemma Collins called out for ‘silly behaviour’ after hugging pony
Gemma Collins has been blasted by fans on social media after she posted a picture on Instagram of herself hugging a free roaming pony in the New Forest
Backlash as Gemma Collins called out for ‘silly behaviour’ after hugging pony
Gemma Collins has been blasted by fans on social media after she posted a picture on Instagram of herself hugging a free roaming pony in the New Forest