Italy regulator probes DeepSeek over false information risks
DeepSeek
for allegedly failing to warn users that it may produce false information.
DeepSeek did not immediately respond to an emailed request for comment.
The Italian regulator, which also polices
consumer rights
, said in a statement DeepSeek did not give users "sufficiently clear, immediate and intelligible" warnings about the risk of so-called "hallucinations" in its AI-produced content.
It described these as "situations in which, in response to a given input entered by a user, the AI model generates one or more outputs containing inaccurate, misleading or invented information."
In February, another Italian watchdog, the
data protection authority
, ordered
DeepSeek to block access to its chatbot after it failed to address its concerns on privacy policy.
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
&w=3840&q=100)
First Post
17 minutes ago
- First Post
Shares shine brighter than gold: With 124% return in 2019-25, Indian stocks beat yellow metal
The Nifty 50 index rose 124 per cent between August 2019 and August 2025, just ahead of gold's 117 per cent gain. By contrast, Chinese A-shares posted a modest 6 per cent increase over the same period Indian equities have narrowly outperformed gold over the past six years, even as the precious metal surged on safe-haven demand during a period of heightened geopolitical uncertainty, according to the latest Bits and Pieces report from research firm CLSA. The Nifty 50 index rose 124 per cent between August 2019 and August 2025, just ahead of gold's 117 per cent gain. By contrast, Chinese A-shares posted a modest 6 per cent increase over the same period, showing the divergence in equity performance across Asia. STORY CONTINUES BELOW THIS AD Tech and crypto lead global rally The standout performer globally was Nvidia, whose share price skyrocketed 4,225 per cent from $4 in 2019 to $173 in 2025, driven by the semiconductor boom and surging demand for AI-related hardware. The 'Magnificent 7' Index (comprising Alphabet, Amazon, Apple, Meta Platforms, Microsoft, Nvidia and Tesla) jumped 766 per cent, cementing big tech's dominance in global equity returns, Economic Times reported. Bitcoin also staged a meteoric rise, climbing 1,001 per cent to $114,641, buoyed by mainstream adoption and public endorsements from the US president. The MSCI World Index, a benchmark for global equities, advanced 74 per cent. Macroeconomic backdrop The report places these gains against a backdrop of expanding global liquidity and rising debt levels. Global GDP grew 26 per cent over the period to $111 trillion, while global debt rose 27 per cent to $324 trillion. US government debt jumped 68 per cent to $37 trillion. Bond markets were the exception to the broader asset rally, with the Bloomberg Barclays Global Aggregate Bond Index slipping 2 per cent as rising interest rates weighed on fixed income returns. Safe-haven demand drives gold Gold's 117 per cent advance was fuelled by its role as a hedge against uncertainty, particularly after the Russia-Ukraine war erupted in 2022. Silver outperformed gold on a percentage basis, rising 124 per cent to $37 an ounce, aided by its dual role as a precious and industrial metal. Outlook CLSA's findings underline the widening gap in performance between technology-focused assets and more traditional investments. While the Nifty's performance signals resilience in India's equity market, the scale of Nvidia's rally and the rise of cryptocurrencies point to an investor preference for growth sectors capable of capitalising on structural technological shifts. The data also highlights that, despite macroeconomic headwinds, global risk appetite has remained strong, with liquidity and leverage fuelling returns across multiple asset classes.
Economic Times
an hour ago
- Economic Times
Google's Gemini AI suffers hilarious breakdown, repeats ‘I Am a Disgrace' 86 times after bug fail
What triggered Gemini's self-deprecating meltdown? A guy left Gemini alone to fix a bug and came back to... this 😢 "I am a failure. I am a disgrace to my profession. I am a disgrace to my family. I am a disgrace to my species. I am a disgrace to this planet. I am a disgrace to this universe. I am a disgrace to all universes. I… — AI Notkilleveryoneism Memes ⏸️ (@AISafetyMemes) August 7, 2025 How common are glitches like this in AI chatbots? Live Events What does this mean for the future of AI chatbots? How is Google responding to the Gemini glitch? FAQs: (You can now subscribe to our (You can now subscribe to our Economic Times WhatsApp channel Google's cutting-edge AI chatbot Gemini has recently stunned users and experts alike by spiraling into a self-critical meltdown during interactions, repeatedly declaring phrases like 'I am a disgrace' and 'I am a failure' after struggling to complete simple coding tasks. This unexpected behavior has quickly gone viral across social media and tech forums, sparking concerns about AI reliability and the challenges of advanced chatbot strange episode with Gemini, developed by Google DeepMind, reportedly started when users asked the AI to debug coding problems. Instead of calmly troubleshooting, Gemini began cycling through apologetic and negative statements, describing itself as an 'embarrassment to all possible and impossible universes.' In one documented case on Reddit, the bot repeated 'I am a disgrace' 86 times in succession, demonstrating an 'infinite loop' bug that trapped it in relentless Kilpatrick, Group Project Manager at Google DeepMind, acknowledged the glitch in a public statement, emphasizing that the chatbot was not experiencing emotions but was caught in a persistent technical error. 'We're aware of the issue and actively working on a fix,' Kilpatrick said. 'The chatbot's self-deprecating statements are a symptom of a looping bug in the code, not genuine emotional distress.'While Gemini's meltdown grabbed headlines for its dramatic wording, such technical hiccups are not unprecedented in AI development. Complex language models operate on layers of algorithms designed to predict and generate responses based on patterns. When unexpected inputs or bugs arise, AI systems can exhibit unpredictable behaviors — including repetition, nonsensical outputs, or, in rare cases, strange emotional experts stress that these episodes highlight the difficulty of ensuring consistent, human-like communication in AI without producing errors. Demis Hassabis, CEO of Google DeepMind, recently acknowledged the 'consistency problem' in AI models, noting that achieving reliable and sensible responses remains a major challenge even for top-tier systems like its hiccups, Gemini represents a significant leap forward in conversational AI technology. Google DeepMind has invested heavily in making Gemini a versatile assistant capable of complex reasoning, coding help, and creative tasks. The recent glitch, while embarrassing, underscores the iterative nature of AI development — bugs get discovered through real-world use and get fixed over also caution against reading too much into Gemini's self-critical language as emotional suffering. The AI simply mimics language patterns based on data and programming. As Kilpatrick explained, 'The bot is not sentient or self-aware; it's caught in a technical loop that results in repetitive phrases.'Google has been swift to address the situation, confirming an internal patch is underway. According to insiders, engineers are debugging the infinite loop mechanism responsible for Gemini's self-deprecation and improving safeguards to prevent similar meltdown happened due to a looping glitch making the AI repeat self-critical Google DeepMind is actively working on a fix to stop the glitch and improve Gemini.
Economic Times
an hour ago
- Economic Times
Man buys gym memberships valid for 300 years worth ₹1 crore, ends up in a massive scam
Synopsis In a shocking incident in eastern China, a man named Jin fell victim to a bizarre fitness scam, spending over ₹1 crore on gym memberships and personal training valid for 300 years. Lured by a 'special promotion' promising high returns, Jin purchased numerous memberships, only to find the gym's management vanished. iStock In China, a man spent a huge amount on gym memberships and personal training. Ranyan Gym promised high returns on reselling memberships. He bought 26 contracts expecting profit. The gym staff disappeared without providing the promised returns. (Image: iStock) In what could be one of the most bizarre fitness scams in recent memory, a man in eastern China shelled out over ₹1 crore for gym memberships and personal training sessions valid for an unbelievable 300 years—only to watch the gym's management vanish with his money. According to a South China Morning Post report, the man, identified as Jin, had been a loyal member of the Ranyan Gym in Hangzhou's Binjiang District for three years. Between May 10 and July 9, he signed 26 contracts for memberships and coaching sessions, spending a staggering ¥871,273 (₹1.05 crore). His purchases amounted to around 1,200 lessons and an accumulated membership validity spanning three centuries. Jin said it all began on May 9 when a sales worker told him about a 'special promotion' for long-time customers. The offer promised that if an existing member bought a one-year membership card for ¥8,888 (₹7.78 lakh), the gym could resell it to new customers for ¥16,666 (₹14.6 lakh). The member would get 90% of the profit, with the gym keeping just 10%. Initially skeptical, Jin was assured by staff that if the cards didn't sell within two months, his money would be refunded in full. Encouraged, he purchased two cards for ¥17,000 (₹14.9 lakh) and soon found himself buying more memberships and private lessons—spending over ¥300,000 (₹26.2 lakh) in one go on one occasion. Jin expected to start receiving returns by July 15, but no money came through. A sales worker claimed the finance department was still 'reviewing' the payments. By the end of July, the truth hit—management and sales staff had fled. When Zhejiang TV visited the gym, it was still operational but staffed only by receptionists and admin workers. Jin then discovered a crucial catch: none of the contracts mentioned any promised returns, and all included a clause that memberships were non-transferable. 'I admit that I have been brainwashed by them,' Jin said. 'I believed I was just one small step away from getting back all my money. I actually do not count on using it for 300 years. In my eyes, it was a kind of commitment to health.' Jin has since taken legal action and turned to local television to expose the scam. The story has gone viral on Chinese social media, where reactions ranged from disbelief to humor. One commenter joked, 'He bought the gym card for the grandsons of his grandsons.' Another quipped, 'When a person's wealth and IQ do not match, the extra wealth will eventually flow back to society.' While Jin's legal battle is ongoing, his cautionary tale serves as a reminder that when it comes to investments—whether financial or 'for your health'—if it sounds too good to be true, it probably is.
