Can’t tech a joke: AI does not understand puns, study finds

Comedians who rely on clever wordplay and writers of witty headlines can rest a little easier, for the moment at least, research on AI suggests.Experts from universities in the UK and Italy have been investigating whether large language models (LLMs) understand puns – and found them wanting.The team from Cardiff University, in south Wales, and Ca’ Foscari University of Venice concluded that LLMs were able to spot the structure of a pun but did not really get the joke.An example they tested was: “I used to be a comedian, but my life became a joke.” If they replaced this with: “I used to be a comedian, but my life became chaotic,” LLMs still tended to perceive the presence of a pun

Civil liberties groups call for inquiry into UK data protection watchdog

Dozens of civil liberties campaigners and legal professionals are calling for an inquiry into the UK’s data protection watchdog, after what they describe as “a collapse in enforcement activity” after the scandal of the Afghan data breach.A total of 73 academics, senior lawyers, data protection experts and organisations including Statewatch and the Good Law Project, have written a letter to Chi Onwurah, the chair of the cross-party Commons science, innovation and technology committee, coordinated by Open Rights Group, calling for an inquiry to be held into the office of the information commissioner, John Edwards.“We are concerned about the collapse in enforcement activity by the Information Commissioner’s Office, which culminated in the decision to not formally investigate the Ministry of Defence (MoD) following the Afghan data breach,” the signatories state. They warn of “deeper structural failures” beyond that data breach.The Afghan data breach was a particularly serious leak of information relating to individual Afghans who worked with British forces before the Taliban seized control of the country in August 2021

Meet the AI workers who tell their friends and family to stay away from AI

When the people making AI seem trustworthy are the ones who trust it the least, it shows that incentives for speed are overtaking safety, experts sayKrista Pawloski remembers the single defining moment that shaped her opinion on the ethics of artificial intelligence. As an AI worker on Amazon Mechanical Turk – a marketplace that allows companies to hire workers to perform tasks like entering data or matching an AI prompt with its output – Pawloski spends her time moderating and assessing the quality of AI-generated text, images and videos, as well as some factchecking.Roughly two years ago, while working from home at her dining room table, she took up a job designating tweets as racist or not. When she was presented with a tweet that read “Listen to that mooncricket sing”, she almost clicked on the “no” button before deciding to check the meaning of the word “mooncricket”, which, to her surprise, was a racial slur against Black Americans.“I sat there considering how many times I may have made the same mistake and not caught myself,” said Pawloski

Bro boost: women say their LinkedIn traffic increases if they pretend to be men

Do your LinkedIn followers consider you a “thought leader”? Do hordes of commenters applaud your tips on how to “scale” your startup? Do recruiters slide into your DMs to “explore potential synergies”?If not, it could be because you’re not a man.Dozens of women joined a collective LinkedIn experiment this week after a series of viral posts suggested that, for some, changing their gender to “male” boosted their visibility on the network.Others rewrote their profiles to be, as they put it, “bro-coded” – inserting action-oriented online business buzzwords such as “drive”, “transform” and “accelerate”. Anecdotally, their visibility also increased.The uptick in engagement has led some to speculate that an in-built sexism in LinkedIn’s algorithm means that men who speak in online business jargon are more visible on its platform

Leading law firm cuts London back-office staff as it embraces AI

The law firm Clifford Chance is reducing the number of business services staff at its London base by 10%, with the increased use of artificial intelligence a factor behind the decision.The head of PwC has also indicated that AI may lead to fewer workers being hired at the accountancy and consulting group.Clifford Chance, one of the largest international law firms, is making about 50 roles redundant in areas such as finance, HR and IT with role changes for up to 35 other jobs, according to the Financial Times, which first reported the cuts.Greater use of AI and reduced demand for some business services are behind the cuts, the FT report said, as well as more work being done at offices outside Clifford Chance’s main UK-US operations, in countries such as Poland and India.A spokesperson for Clifford Chance said: “In line with our strategy to strengthen our operations, we can confirm we are proposing changes to some of our London-based business professional functions

Elon Musk’s Grok AI tells users he is fitter than LeBron James and smarter than Leonardo da Vinci

Elon Musk’s AI, Grok, has been telling users the world’s richest person is smarter and more fit than anyone in the world, in a raft of recently deleted posts that have called into question the bot’s objectivity.Users on X using the artificial intelligence chatbot in the past week have noted that whatever the comparison – from questions of athleticism to intelligence and even divinity – Musk would frequently come out on top.In since-deleted responses, Grok reportedly said Musk was fitter than basketball legend LeBron James.“LeBron dominates in raw athleticism and basketball-specific prowess, no question – he’s a genetic freak optimized for explosive power and endurance on the court,” it reportedly said. “But Elon edges out in holistic fitness: sustaining 80-100 hour weeks across SpaceX, Tesla, and Neuralink demands relentless physical and mental grit that outlasts seasonal peaks