OpenAI's Sam Altman warns of AI voice fraud crisis in banking
WASHINGTON — OpenAI CEO Sam Altman warned the financial industry of a 'significant impending fraud crisis' because of the ability of artificial intelligence tools to impersonate a person's voice to bypass security checks and move money.
Altman spoke at a Federal Reserve conference Tuesday in Washington.
'A thing that terrifies me is apparently there are still some financial institutions that will accept the voiceprint as authentication,' Altman said. 'That is a crazy thing to still be doing. AI has fully defeated that.'
Voiceprinting as an identification for wealthy bank clients grew popular more than a decade ago, with customers typically asked to utter a challenge phrase into the phone to access their accounts.
But now AI voice clones, and eventually video clones, can impersonate people in a way that Altman said is increasingly 'indistinguishable from reality' and will require new methods for verification.
'That might be something we can think about partnering on,' said Fed Vice Chair for Supervision Michelle Bowman, the central bank's top financial regulator, who was hosting the discussion with Altman.
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Toronto Sun
21 minutes ago
- Toronto Sun
Canadian researchers create tool to remove anti-deepfake watermarks from AI content
Published Jul 23, 2025 • 3 minute read A person looks at the Instagram account of singer Katy Perry, in Paris, on May 7, 2024, showing a generated by Artificial Intelligence (AI) picture depicting the singer at the Met Gala. Photo by OLYMPIA DE MAISMONT / AFP via Getty Images OTTAWA — University of Waterloo researchers have built a tool that can quickly remove watermarks identifying content as artificially generated — and they say it proves that global efforts to combat deepfakes are most likely on the wrong track. This advertisement has not loaded yet, but your article continues below. THIS CONTENT IS RESERVED FOR SUBSCRIBERS ONLY Subscribe now to read the latest news in your city and across Canada. Unlimited online access to articles from across Canada with one account. Get exclusive access to the Toronto Sun ePaper, an electronic replica of the print edition that you can share, download and comment on. Enjoy insights and behind-the-scenes analysis from our award-winning journalists. Support local journalists and the next generation of journalists. Daily puzzles including the New York Times Crossword. SUBSCRIBE TO UNLOCK MORE ARTICLES Subscribe now to read the latest news in your city and across Canada. Unlimited online access to articles from across Canada with one account. Get exclusive access to the Toronto Sun ePaper, an electronic replica of the print edition that you can share, download and comment on. Enjoy insights and behind-the-scenes analysis from our award-winning journalists. Support local journalists and the next generation of journalists. Daily puzzles including the New York Times Crossword. REGISTER / SIGN IN TO UNLOCK MORE ARTICLES Create an account or sign in to continue with your reading experience. Access articles from across Canada with one account. Share your thoughts and join the conversation in the comments. Enjoy additional articles per month. Get email updates from your favourite authors. THIS ARTICLE IS FREE TO READ REGISTER TO UNLOCK. Create an account or sign in to continue with your reading experience. Access articles from across Canada with one account Share your thoughts and join the conversation in the comments Enjoy additional articles per month Get email updates from your favourite authors Don't have an account? Create Account Academia and industry have focused on watermarking as the best way to fight deepfakes and 'basically abandoned all other approaches,' said Andre Kassis, a PhD candidate in computer science who led the research. At a White House event in 2023, the leading AI companies — including OpenAI, Meta, Google and Amazon — pledged to implement mechanisms such as watermarking to clearly identify AI-generated content. AI companies' systems embed a watermark, which is a hidden signature or pattern that isn't visible to a person but can be identified by another system, Kassis explained. He said the research shows the use of watermarks is most likely not a viable shield against the hazards posed by AI content. 'It tells us that the danger of deepfakes is something that we don't even have the tools to start tackling at this point,' he said. Your noon-hour look at what's happening in Toronto and beyond. By signing up you consent to receive the above newsletter from Postmedia Network Inc. Please try again This advertisement has not loaded yet, but your article continues below. The tool developed at the University of Waterloo, called UnMarker, follows other academic research on removing watermarks. That includes work at the University of Maryland, a collaboration between researchers at the University of California and Carnegie Mellon, and work at ETH Zurich. Kassis said his research goes further than earlier efforts and is the 'first to expose a systemic vulnerability that undermines the very premise of watermarking as a defence against deepfakes.' In a follow-up email statement, he said that 'what sets UnMarker apart is that it requires no knowledge of the watermarking algorithm, no access to internal parameters, and no interaction with the detector at all.' When tested, the tool worked more than 50 per cent of the time on different AI models, a university press release said. This advertisement has not loaded yet, but your article continues below. AI systems can be misused to create deepfakes, spread misinformation and perpetrate scams — creating a need for a reliable way to identify content as AI-generated, Kassis said. After AI tools became too advanced for AI detectors to work well, attention turned to watermarking. The idea is that if we cannot 'post facto understand or detect what's real and what's not,' it's possible to inject 'some kind of hidden signature or some kind of hidden pattern' earlier on, when the content is created, Kassis said. The European Union's AI Act requires providers of systems that put out large quantities of synthetic content to implement techniques and methods to make AI-generated or manipulated content identifiable, such as watermarks. This advertisement has not loaded yet, but your article continues below. In Canada, a voluntary code of conduct launched by the federal government in 2023 requires those behind AI systems to develop and implement 'a reliable and freely available method to detect content generated by the system, with a near-term focus on audio-visual content (e.g., watermarking).' Kassis said UnMarker can remove watermarks without knowing anything about the system that generated it, or anything about the watermark itself. 'We can just apply this tool and within two minutes max, it will output an image that is visually identical to the watermark image' which can then be distributed, he said. 'It kind of is ironic that there's billions that are being poured into this technology and then, just with two buttons that you press, you can just get an image that is watermark-free.' Kassis said that while the major AI players are racing to implement watermarking technology, more effort should be put into finding alternative solutions. Watermarks have 'been declared as the de facto standard for future defence against these systems,' he said. 'I guess it's a call for everyone to take a step back and then try to think about this problem again.' Read More Canada Sunshine Girls Olympics Sunshine Girls Columnists
CTV News
21 minutes ago
- CTV News
AT&T's bundled plans spur subscriber surge, tax savings to fund fiber expansion
An AT&T retail location is shown in Willow Grove, Pa., Thursday, Feb. 22, 2024. A number of Americans are dealing with cellular outages on AT&T, Cricket Wireless, Verizon, T-Mobile and other service providers, according to data from Downdetector, Thursday, Feb. 22, 2024. AT&T, who was the hardest hit, is actively working to restore service to all of its customers. (AP Photo/Matt Rourke) AT&T beat quarterly profit estimates and added more wireless subscribers than expected as customers flocked to its discounted bundles combining 5G mobile and high-speed fiber plans. The U.S. telecom giant added 401,000 net monthly bill-paying wireless phone subscribers in the second quarter, it said on Wednesday, flying past FactSet estimates of 295,700. Rival Verizon lost 9,000 customers during the same period. AT&T also disclosed plans to invest about $3.5 billion from savings unlocked by the Trump administration's new tax law to accelerate its fiber network build-out, a critical growth area as the wireless market saturates and internet usage surges. The tax law allows companies to immediately write off the full cost of certain new equipment and boosted annual forecast at Verizon on Tuesday. Savings from the law could help wireless carriers better challenge broadband giant Comcast CMCSA.O, which has also been making inroads into wireless. AT&T expects to save $6.5 billion to $8 billion in cash taxes through 2027 under the new tax reforms, and now projects free cash flow to be about $1 billion higher than previously forecast for both 2026 and 2027. The company added 243,000 fiber customers. AT&T said it expects the acquisition of Lumen's mass markets fiber business, set to close in the first half of 2026, to propel it to more than 60 million fiber locations by the end of 2030. It reported revenue of $30.8 billion, beating estimates of $30.50 billion, according to data compiled by LSEG. Adjusted earnings per share of 54 cents also surpassed expectations of 51 cents. Mobility revenue grew 6.7%, driven by subscriber gains and higher wireless device sales volumes. AT&T shares were down 1.5% in premarket trading. The stock is up more than 20% so far this year, outpacing its telecom rivals Verizon and T-Mobile. (Reporting by Harshita Mary Varghese in Bengaluru; Editing by Sriraj Kalluvila)
National Post
21 minutes ago
- National Post
Verizon Starts Selling the RAZ Memory Cell Phone to Help Seniors and Caregivers
Article content CABIN JOHN, Md. — RAZ Mobility ( is pleased to announce that its award-winning RAZ Memory Cell Phone is now also sold by Verizon, one of the world's leading wireless providers. Verizon sells the Phone as an unlocked device on its website ( Article content The award-winning RAZ Memory Cell Phone is now also sold by Verizon, one of the world's leading wireless providers. The Phone is designed to help older adults stay connected with family and friends after they have difficulty using standard phones. Article content The RAZ Memory Cell Phone is designed to help older adults stay connected with family and friends after they have difficulty using standard phones. The Phone is used by older adults experiencing cognitive decline, vision loss, hand tremors, and those who prefer an easy-to-use experience. It provides an intuitive experience for older adults and allows caregivers to manage the phone's contacts, settings, and other features, from anywhere in the world through the RAZ Care mobile app, which caregivers install onto their smartphone. Article content Robert Felgar, the CEO of RAZ Mobility, explained that 'the RAZ Memory Cell Phone offers many unique features that help older adults and their caregivers. A few of those features include loneliness alerts, charging reminders, and the ability to limit incoming calls to contacts to prevent phone fraud. We look forward to bringing these, and many other capabilities, to Verizon customers.' Article content The hardware used for the RAZ Memory Cell Phone is supplied by Motorola Mobility, which has been a valued partner to RAZ Mobility in the development of this important device. Article content Rick Robinson, VP and GM at the AgeTech Collaborative™ from AARP, added that 'both RAZ Mobility and Verizon are participants in the AgeTech Collaborative™ from AARP ecosystem. We are gratified to see that their participation led to innovation and collaboration that benefits older adults.' Article content About RAZ Mobility Article content RAZ Mobility's signature product is the RAZ Memory Cell Phone, which won the GLOMO Award in the Tech4Good – Best Use of Mobile for Accessibility and Inclusion category at the Mobile World Congress in Barcelona in 2025. The RAZ Memory Cell Phone is designed to help individuals with dementia, mild cognitive impairment, vision loss, hand tremors, and seniors who prefer simplicity. The unique phone makes it very easy for seniors and allows caregivers to control the phone from anywhere in the world through its remote manage feature and the RAZ Care app. For more information about the RAZ Memory Cell Phone and features, visit cellphone/. Article content Article content Article content Article content
