
There's a '10% to 20% chance' that AI will displace humans completely, says 'godfather' of the technology
Geoffrey Hinton, the computer scientist known as a "Godfather of AI," says the technology he helped create is getting increasingly scary — and not enough people are taking the risks of artificial intelligence seriously.
"There's risks that come from people misusing AI, and that's most of the risks and all of the short-term risks. And then there's risks that come from AI getting super smart and understanding it doesn't need us," Hinton, an ex-Google vice president who won the 2018 Turing Award for his decades of pioneering work on AI and deep learning, said on Monday's podcast episode of "The Diary of a CEO."
AI's rapid spread across the world includes a rising number of students using ChatGPT, CEOs mandating the technology's use in their workplaces and tech luminaries like Mark Cuban and Jensen Huang saying that AI will soon be the differentiator between success and failure, for employees and businesses alike.
But the engineers who build today's AI systems still don't fully understand how the technology works and evolves, leaving many of them split on its future. Some predict a future technological uprising where AIs displace humans, and others dismiss the worry as science fiction, Hinton said.
"I think both of those positions are extreme," said Hinton, 77. "I often say [there's a] 10% to 20% chance [for AI] to wipe us out. But that's just gut, based on the idea that we're still making them and we're pretty ingenious. And the hope is that if enough smart people do enough research with enough resources, we'll figure out a way to build them so they'll never want to harm us."As for the shorter-term risks that Hinton described, AI suffers from "hallucinations" — factual inaccuracies seemingly created from thin air — and allows people to manufacture fake images, videos and audio with relative ease. The technology also seems poised to automate a series of entry-level job responsibilities in many white-collar industries, though some tech leaders say it'll ultimately create more jobs than it replaces.
And AI-enabled scams are becoming increasingly common, CoinStructive co-founders Chris Groshong and Joseph Albiñana told Forbes on June 12. Scammers can use the tech to create lip-synced conversations, fake firmware and impersonate other people in video chats, where unassuming victims share personal and financial information, said Groshong, whose company is a crypto compliance and investigation firm.
Humanity is likely at "a kind of turning point," Hinton said on CBS' "60 Minutes," in an interview that first aired on Oct. 8, 2023. "I think my main message is there's enormous uncertainty about what's going to happen next."
Hinton quit his job at Google in May 2023 after a decade with the company so he could speak freely about the risks posed by AI, he told The New York Times at the time. But the worst-case scenario is no sure thing, and industries like health care have already benefitted tremendously from AI, he said on "60 Minutes."
During that television appearance, he called for more research to understand AI, government regulations to rein in the technology and worldwide bans on AI-powered military robots. Whatever AI guardrails get put into place — whether by tech companies, the U.S. federal government or other governments across the world — they need to happen soon, Hinton said.
On Monday's podcast interview, Hinton offered listeners advice for safeguarding themselves against AI-augmented scams: Diversify where your money is held, and regularly back up your data to an external hard drive. If you fall victim to a scam, you'll only lose some of your money, and you can quickly restore your computer and phone to working order, he said.
Hinton holds his own money in three separate banks, he said.
More generally, diversifying your assets is a smart financial move, according to financial analyst and personal finance expert Chris Browning. He recommends spreading your money across five different accounts: two separate checking accounts for bills and lifestyle spending, and three savings accounts for long-term savings, short-term savings and an emergency fund, he wrote for CNBC Make It on December 9, 2021.
"Try not to keep all your accounts at the same bank," wrote Browning. "In case technology fails at one institution, for example, you have accounts at other banks to fall back on."
Hashtags

Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles


Tom's Guide
31 minutes ago
- Tom's Guide
This is the one upgrade that will get me to switch my Pixel Watch 3 for a Pixel Watch 4
I've always had very mixed feelings about smartwatches, but finally caved and picked up the Google Pixel Watch 3 last year. While my feelings haven't completely turned around, I have managed to fill a few niches with the wearable — but there's one thing I'd like to change. If you remember the original Pixel Watch, which was released back in 2021, you'll be aware that it came with a wireless charging-like system similar to the Apple Watch 10 and Galaxy Watch 7. By the time the Pixel Watch 2 came around, though, that feature was gone. And having used the replacement charging system and logged the problems it seems to have, I wish the Pixel Watch 4 would bring back something similar to the old arrangement — even though I know it's not going to happen should the Pixel 4 show up as expected later this year alongside the Pixel 10 smartphone series. Rather than having a wireless charging system, both the Pixel Watch 2 and Watch 3 draw their power from a 4-pin connector. It's the same charging system as recent Fitbit watches, like the Fitbit Sense 2 and Versa 4 — to the point where the chargers are (just about) compatible with each other. But while Fitbit watches are now functionally dead, the charging connection is still very much alive. The problem is, it doesn't work nearly as well on my Pixel Watch 3 as it did on my Fitbit Versa 4. I never actually had an issue with getting power into that watch, but the Pixel Watch 3 doesn't seem to want to play by the same rules. I've lost count of the number of times I've put my watch down to charge, and it just sat there not taking in power. I've done everything from making sure the pins are perfectly aligned, that the connectors are all clean, and there will still be times when the watch essentially goes, "hmm, nah." I never had that issue with the Versa 4, and that wearable was considerably cheaper — not to mention less advanced. Aside from the fact the charging pins liked scratching the back of the watch, it always recharged when I needed it to. I don't know why that's the case, but it's incredibly frustrating — and I'd love to be able to slap my watch onto a charging puck haphazardly with the confidence that it won't just sit there doing nothing. Plus, it would be nice to use my the reverse wireless charging on my Pixel 9 Pro to juice up my watch. When you look at the reasons why Google scrapped wireless charging for Pixel Watch 2 and 3, the reasons make sense. The main benefits are that pin charging offers faster charging speeds, which is helpful when you need to recharge your watch every day. This charging system also produces significantly less heat. In fact the heat problem was so bad that Google had to put out an update to nerf the charging speed to keep the temperature down to the point where a 100% recharge took almost 2 hours to finish — which is not good. I had wondered whether thickness was another factor, since it always seems to be with smartphones, but that doesn't seem to be the case. The Pixel Watch and its successors are all 12.3mm thick; in fact, they all look pretty similar, though admittedly, the original Pixel Watch is a lot rounder at the bottom compared to its younger siblings. That said, removing the charging coils from the wearable could have made space for other things. The Pixel Watch 2 has sensors for EDA, skin temperature, irregular heart beat rhythm and an improved multi-point heart rate sensor. Not to mention, it offers a larger battery than its predecessor. It was never officially confirmed that these features required the charging coil space, but I'd bet money Google definitely put the extra real estate to good use once the charging pins were decided on. As problematic as the Pixel Watch's wireless charging was, I'd still like to see some kind of puck-charging system come back on Pixel Watch 4 — or on some future model if it's too late to do anything about this year's offering. Google will need to work out the kinks, and improve wireless charging speeds without turning the wearable into a miniature burn risk in the process. Or, you know, at least make the 4-pin charging more consistent than it currently is. It really should not be hit-and-miss the way it has been for me, because like any charger, it should just be able to work. Hopefully, that isn't just a pipe dream.


Business Journals
33 minutes ago
- Business Journals
49% of CRE leaders want more AI – Here's why
The commercial real estate market still faces many challenges and executives are looking for ways to lower costs — many are turning to technology. In Citrin Cooperman's latest survey of 440 commercial real estate leaders in conjunction with the Commercial Observer, respondents said AI was their number one opportunity over the next 12 months. They ranked it above new property management systems, better cybersecurity, more process automation, solar, listing services, and all else. Though as we will explore, AI has potential to help in nearly all of these areas as well. Using AI to model building information The No. 1 AI use case, 34% of respondents Leaders say the most common use for AI right now is modeling building information. In some cases that means working with geospatial information, such as using AI to visualize floorplates or run feasibility studies. A feasibility study provider can now use software to generate dozens of options for consideration in days rather than months. But modeling can also mean financial modeling, such as running the numbers on what it will take for a property or portfolio to achieve NOI. Current AI technologies can also help corporate development teams determine where buildings should go. The AI platform SiteZeus can ingest cell phone, traffic, parking, real estate, walkability, weather, and other data sources to show teams on a map the exact spots they should select. Some franchisors are providing this tool to their franchisees. AI cannot do everything, however. At least not without assistance. It cannot, for example, figure out how to actually put those real estate deals together. It cannot validate those decisions, run studies on the revenue or tax impact implications, nor advise on which financial instrument to use. To discuss how to turn the information generated by a real estate AI tool into reports, decks, and models your team can make decisions on, please reach out to Citrin Cooperman's Digital Services Practice. Using AI to source and match deals The No. 2 AI use case, 20% of respondents Teams can use AI to source and sort through deals. AI tools can scan all possible opportunities and rank them by the commercial team's requirements. Typically, commercial teams lose a great deal of time on market research, outreach, screening, and due diligence. AI tools are one way for them to reclaim and repurpose it. The software Skyline AI by JLL, for example, provides real estate investors with a data-driven view of their market that goes much deeper than typical census data, with a view of factors such as opportunity zones. offers automated commercial property appraisals and automatic document review, so brokers can save hours on reviewing leases, loan agreements, operating agreements, and more. Tools like Vena or Power BI can also reveal deal insights within your existing data: Your team can use them to create smart dashboards to forecast continuously without requesting new reports. AI tools do have one weakness: They are only as good as the data within your ERP system, which most real estate firms have not updated in years. A Citrin Cooperman survey of 1,000 private companies found that 42% of real estate companies say their number one blocker to adopting AI is 'integrations with existing enterprise resource planning (ERP) systems.' Our Digital Services Practice can review your ERP system to understand if your workflows and data are set up for you to use in deal-sorting AI. Property management The No. 3 use case, 20% of respondents Commercial real estate companies are using AI to directly manage their properties, plugging it into property management software to spot trends. This allows managers to see if vacancies are subtly rising so they know to invest in more marketing or start offering concessions. Similarly, they can spot fluctuations in utility and repair costs. Some smart devices like Omnidian for solar or Checkit for commercial appliances can detect near-imperceptible shifts in electrical device power signals to understand when they need maintenance — and auto-dispatch a technician for preventative maintenance. The 'smart' property trend can also be smart for tax reasons — investing in your building's service-layer can help you improve the amenities and take advantage of tax credits. Smart and environmentally rated devices may pay for themselves. More smart-devices and self-serve check-in kiosks and building management apps can also mean less need for staff walking the floors. This does not always mean fewer employees, but it does mean shifting those roles to other departments and may require you to review your union contracts and local employment regulations. Finance The No. 4 use case, 9% of respondents Finally, commercial real estate teams are using AI to complete finance tasks. Many of these AI systems are coming into the real estate world through general applications, such as how Google's Gemini assistant is now available in Google Sheets. Whereas some are specific to real estate. The software Yardi offers a type of smart spreadsheet reporting that Citrin Cooperman Managing Partner of Business Process Outsourcing Mike Zyborowicz calls the Swiss Army Knife of property management and notes that, 'It enables you to do everything from managing your assets to crunching numbers so you can better manage tenant satisfaction or overhead expenses.' What AI tools will not do, however, is certify the results of studies around how to increase occupancy and improve company profitability. Nor can AI tools yet look forward to advise on where the market is headed and suggest new configurations for your capital stack, such as preferred equity or mezzanine debt. AI works best when supported by a financial advisor AI has vast potential in the commercial real estate sector and is one sure way to help reduce costs. But it is never as easy as 'plug and play.' To turn your team's AI pilots into actionable insights your firm can place class-A bets on, you may need a financial firm that specializes in advisory, audit, assurance, tax, and technology. Only then can you ensure the data within your systems is worthy of basing decisions on, and that you have considered the full breadth of financial and tax implications. To discuss how Citrin Cooperman's Real Estate Industry Practice can help your business reduce operating expenses with investments in AI, and maximize the tax benefits of such investments, reach out to Jessica Garber or Adam Lazarus. Citrin Cooperman is one of the nation's largest and fastest-growing professional services firms. Since 1979 our daily mission has been to help middle-market companies and high net worth individuals find success in their business and personal financial lives through our proactive guidance, specialized services, and passion for excellence. Rooted in our core values, we deliver a comprehensive, integrated business approach, including tailored insights throughout the lifecycle of our clients. Whether your operations and assets are located around the corner or across the globe, we provide new perspectives on strategies that help you achieve your short- and long-term goals. With over 30 offices and more than 3,500 professionals, Citrin Cooperman is included in the Top 20 Firms by Accounting Today. Learn more about Citrin Cooperman at "Citrin Cooperman" is the brand under which Citrin Cooperman & Company, LLP, a licensed independent CPA firm, and Citrin Cooperman Advisors LLC serve clients' business needs. The two firms operate as separate legal entities in an alternative practice structure. The entities of Citrin Cooperman & Company, LLP and Citrin Cooperman Advisors LLC are independent member firms of the Moore North America, Inc. (MNA) Association, which is itself a regional member of Moore Global Network Limited (MGNL). All the firms associated with MNA are independently owned and managed entities. Their membership in, or association with, MNA should not be construed as constituting or implying any partnership between them.
Yahoo
42 minutes ago
- Yahoo
Zango AI Emerges From Stealth, Raising $4.8m to Revolutionise Financial Compliance Systems With AI Agents
LONDON, July 01, 2025--(BUSINESS WIRE)--London-based Zango has raised $4.8m, in a round led by Nexus Venture Partners, a leading global venture capital firm with over $2.6 billion capital under management. Other participants in the latest round include South Park Commons, Richard Davies, CEO of Allica Bank, Alan Morgan, former head of Financial Services at McKinsey (EMEA), Mark Ransford, Notion Capital, No Label Ventures and Start Ventures. Ritesh Singhania, Co-founder of Zango says, "Traditional SaaS stops at KYC and AML, leaving the rest of compliance buried in spreadsheets and manual reviews. Zango changes that. Our regulation-aware AI agents continuously track regulatory updates (horizon scanning), identify compliance gaps in real time, and keep firms audit-ready — no operational drag, no surprises." Zango is actively used by established banks such as Novobanco, the fourth-largest bank in Portugal, and is gaining traction with leading neo banks in the EU and UK, including Monzo and Juni. Today, financial services regulations stretch across tens of thousands of pages, creating slow, manual workflows and significant costs for businesses striving to stay compliant. Zango addresses this critical challenge by transforming regulatory complexity into a clear list of requirements. Its innovative solution is training regulation-specific Large Language Models (LLMs) to make compliance as intuitive as a simple query, enabling businesses to launch new products and expand into new markets with unprecedented speed and confidence, helping compliance be a growth centre rather than just a cost centre. With compliance becomes as simple as a Google query: "I want to launch a lending product in [Y] market — what do I need to do?" Zango's AI agents don't just read and interpret regulations with a high degree of accuracy — they integrate into a company's day-to-day operations. In one example, a workflow with one of the banks involving the regulator was reduced from 48 hours to under 4 hours, showcasing the power of AI in regulated environments. Compliance becomes a continuous, always-on process rather than a one-time checklist. Traditional consultants give you a green light at a moment in time. But fast-growing businesses change quickly. Six months later, that "green light" is irrelevant — and by the time a regulator shows up, you could be out of bounds. "We don't sell a platform — we sell a solution," said Ritesh Singhania, Co-founder of Zango. "Our AI agents are paired with humans-in-the-loop to ensure 100% accuracy. Peace of mind doesn't come from a tool; it comes from a result. That's why we win against consultants — because they don't just sell software, and neither do we." Zango was co-founded by Ritesh Singhania and Shashank Agarwal, both second-time founders with deep experience in regulatory technology. Ritesh previously founded ClearGlass, a pension compliance platform, and served as Head of Technology at Simplitium (acquired by NASDAQ). Shashank co-founded Third Watch, an AI-powered fraud detection startup (acquired by Razorpay, valued at $7.5bn+), and led trust and compliance engineering at PhonePe, which is gearing up for India's largest IPO. "The global regulatory landscape is ripe for disruption," said Anand Datta, Partner at Nexus Venture Partners. "Ritesh and Shashank, with their firsthand, proven expertise, developed Zango's first-principles approach: uniquely marrying cutting-edge AI with human compliance expertise. Their AI-led solution is already augmenting the compliance team and increasing their efficiency at global financial institutions. We're incredibly excited to be part of their journey." "We are thrilled with how Zango AI has transformed our compliance processes, saving us hundreds of hours of manual repetitive work and allowing our team to focus on strategic initiatives," said Lara Reis, Head of regulatory affairs, Novobanco. "For Horizon scanning, Zango is a game-changer. The platform's current and future modules are set to revolutionize the second line of defense work in regulated industries like ours," Sabina Ausfelt, Head of compliance, Juni. The funds raised will be used to expand teams in London & Bengaluru along with building out the other product modules for an AI-native GRC. Zango will expand into other verticals of financial services as well outside of banking to insurance and asset management. About Zango AI Zango is a regulatory compliance solution that leverages AI alongside trusted subject-matter experts to automate horizon scanning, gap analysis, and controls testing for global financial institutions. Founded in 2024 by experts in AI and financial regulation, the company operates globally with offices in London, San Francisco and Bengaluru. Zango is trusted by enterprise customers navigating complex regulatory environments across the UK, EU, and beyond. The platform is SOC 2 Type II and ISO 27001 certified, with security and scalability built in from the ground up. For more information, visit About Nexus Venture Partners Nexus Venture Partners is a leading early-stage venture capital firm partnering with extraordinary entrepreneurs building product-first companies. With $2.6 billion under management, Nexus operates as a unified team across the US and India. Nexus's portfolio includes Bolt, Postman, Daloopa, Fingerprint, Zepto, Iambic, Infra Market, MinIO, Pubmatic, Rapido, Rancher, Turtlemint, Uniqus and Delhivery. View source version on Contacts Media Contact: Ben Goldsmithben@ +44(0)7788295321 Error in retrieving data Sign in to access your portfolio Error in retrieving data Error in retrieving data Error in retrieving data Error in retrieving data