Google has a new tool to help detect AI-generated content
Google announced a new SynthID Detector tool at Google I/O that lets you check if content has been made with the assistance of Google's AI tools.
In a blog post, Google DeepMind's Pushmeet Kohli describes SynthID Detector as 'a verification portal' that can 'quickly and efficiently identify AI-generated content made with Google AI.' It's also able to 'highlight which parts of the content are more likely to have been watermarked with SynthID.'
SynthID watermarks are applied to AI-generated images, text, audio, and videos, including content generated by Google's Gemini, Imagen, Lyria, and Veo models, Kohli says.
Here's how the tool works, according to Kohli:
When you upload an image, audio track, video or piece of text created using Google's AI tools, the portal will scan the media for a SynthID watermark. If a watermark is detected, the portal will highlight specific portions of the content most likely to be watermarked.
For audio, the portal pinpoints specific segments where a SynthID watermark is detected, and for images, it indicates areas where a watermark is most likely.
Google is starting to roll out the tool to 'early testers,' Kohli says in the post.
'Following the initial testing phase, the portal will gradually be rolled out to users who sign up to the waitlist to gain access to the SynthID Detector,' Kohli tells The Verge. 'We will take learnings from this cohort of professionals and work to implement content transparency more broadly.'
I'm on the waitlist, but I haven't tested the tool myself, so I can't vouch for how well it might work. And will people actually use it when it's widely available? I hope so, but we'll have to wait and see.
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Forbes
40 minutes ago
- Forbes
Bitcoin Suddenly Braced For $7 Trillion ‘Critical' Price Tipping Point After Stark BlackRock Warning
06/01 update below. This post was originally published on May 30 Bitcoin has fallen sharply since hitting an all-time high last week, down 6% despite the U.S. vice president JD Vance issuing a huge bitcoin prediction. Front-run Donald Trump, the White House and Wall Street by subscribing now to Forbes' CryptoAsset & Blockchain Advisor where you can "uncover blockchain blockbusters poised for 1,000% plus gains!" The bitcoin price has rocketed over the last two years, turbo-charged by Wall Street adoption led by the world's largest asset manager BlackRock and then U.S. president Donald Trump's embrace of the technology (with Elon Musk quietly plotting what could be a bitcoin price game-changer). Now, as fears swirl the U.S. dollar could be teetering on the verge of collapse, a serious BlackRock warning that quantum computing could pose an existential risk to bitcoin has been escalated by a Google research paper that found encryption-breaking quantum computers could be a lot closer than previously thought. Sign up now for the free CryptoCodex—A daily five-minute newsletter for traders, investors and the crypto-curious that will get you up to date and keep you ahead of the bitcoin and crypto market bull run BlackRock chief executive Larry Fink has emerged as one of the most bullish voices supporting ... More bitcoin on Wall Street, helping the bitcoin price boom. 'This is a 20-fold decrease in the number of qubits from our previous estimate,' Google Quantum AI researcher Craig Gidney wrote, referring to the number of quantum computer qubits needed to break a public-key encryption algorithm similar to that used by bitcoin. 'If this is even remotely true, combined with everything else happening right now, the only safe trade are hard assets and, dare I say, gold,' investor Chamath Palihapitiya, a vocal supporter of bitcoin who claims to have first bought some when the bitcoin price was just $100, posted to X in response to the paper. 06/01 update: The bitcoin price has now plummeted by almost 10% from its all-time high of $112,000 per bitcoin, falling to just over $103,000 and fast approaching the closely-watched $100,000 level. "Next week will be critical for both the crypto market," analysts with 10x Research led by Markus Thielen wrote in an emailed note that suggested long-term bitcoin holders are beginning to sell. This coming week will see slew of economic and labor market data—topped by Friday's monthly U.S. jobs report—as well as U.S. president Donald Trump's controversial "big, beautiful" tax bill coming to the Senate and legal back-and-forth expected over Trump's market-crashing barrage of global trade tariffs. Meanwhile, bullish bitcoin price analysts are predicting small changes in 'sentiment' could trigger outsized market swings, with huge volumes of cash still sitting on the sidelines. 'With roughly $7 trillion still parked in money market funds and another $2 trillion in fixed income ETFs, even a modest shift in risk sentiment could redirect meaningful capital into crypto and other high-beta assets,' Matt Mena, research strategist at crypto exchange-traded fund (ETF) company 21Shares, said in emailed comments. 'If bitcoin breaks out of the $105,000 to $110,000 range with conviction, we could see a sharp move to $120,000 and, more importantly, reach our previously year-end price target of $138,500 per bitcoin by the end of the summer.' Earlier this month, BlackRock quietly added a serious warning about quantum computing to the list of risks to its huge spot bitcoin exchange-traded fund (ETF). BlackRock, which manages after around $10 trillion worth of assets for investors, spearheaded Wall Street's campaign to bring a long-awaited spot bitcoin ETF to market in 2023, with a fleet of funds debuting in January 2024. The fund now holds around 3% of the 21 million bitcoin that will ever exist, worth $70 billion at the current bitcoin price, which some have warned could be giving BlackRock outsized control over the network. Sign up now for CryptoCodex—A free, daily newsletter for the crypto-curious The bitcoin price has rocketed higher, hitting an all-time high of around $112,000 per bitcoin last ... More week before dropping back. 'If quantum computing technology is able to advance […] it could potentially undermine the viability of many of the cryptographic algorithms used across the world's information technology infrastructure, including the cryptographic algorithms used for digital assets like bitcoin,' BlackRock's amended regulatory filing for its bitcoin fund read. The quantum computing risk to bitcoin and other cryptocurrencies has exploded recently, with tech giants including Google making strides in quantum computing research. 'At this point, no blockchain is ready to withstand a quantum attack when this becomes possible, which could very well be much earlier than 2030,' David Carvalho, the chief executive of decentralized post-quantum infrastructure blockchain Naoris Protocol, said in earlier emailed comments that warned bitcoin is 'sleepwalking into a disaster.'
Yahoo
an hour ago
- Yahoo
Intel's lackluster Arrow Lake appears to have a refresh inbound — Arrow Lake Refresh appears in reference document
When you buy through links on our articles, Future and its syndication partners may earn a commission. If we are to believe a leaked manual for an upcoming Intel W880 motherboard, the Intel Arrow Lake-S processor will seemingly have a refresh successor. Popular hardware leaker momomo_us shared an image of the purported motherboard reference document with the CPU labeled Arrow Lake S/Arrow Lake S Refresh. Previous rumors point to another generation of the Arrow Lake architecture, focusing only on high-end K and KF chips. So, it would make sense for motherboards designed for high-end workstations to take advantage of this refresh. Unfortunately, we do not see any other information on the document, so we will have to wait and see if manufacturers of more mainstream consumer and enthusiast motherboards will release new models for this alleged Arrow Lake update. The Intel Core Ultra 200S series arrived on the scene in October 2024, around two months after the company released its disastrous financial report. Team Blue sorely needed a win to help it get back on course—unfortunately, the Arrow Lake wasn't it, as sales for its latest processor stagnated just a week after its launch. After half a year on the market, the company still struggles to sell its AI chips, with many customers preferring the older Raptor Lake CPUs. One reason behind this could be the chip's lack of gaming prowess. It has consistently trailed behind AMD's outstanding offerings, especially the Ryzen 7 9800X3D, and was even outperformed by Intel's own last-generation chips. And even though the Arrow Lake processors boast an NPU and offer higher AI processing power, most buyers don't care about that. This, combined with the higher prices, meant that there was low demand for these CPUs. If the Arrow Lake-S Refresh chips are indeed arriving, Intel must address the gaming performance issues. Otherwise, buyers would shun them and instead flock towards options that deliver more FPS and do not bottleneck other components. Despite all the negativity in the gaming space for the Arrow Lake-S chips, Intel has made some interesting innovations in the processor. It has improved power consumption and efficiency, reducing its cooling requirements. It's also a productivity powerhouse, with the Intel Core Ultra 9 285K beating most of its competitors in Cinebench. This was the first Intel desktop CPU to use a chiplet-based design, so it's bound to have first-generation hiccups. Hopefully, the company can fix some of the concerns with the refresh, allowing us to get the most out of the Arrow Lake architecture. Follow Tom's Hardware on Google News to get our up-to-date news, analysis, and reviews in your feeds. Make sure to click the Follow button.
Forbes
an hour ago
- Forbes
AI May Be Erasing Entry-Level Jobs
Purple ID card holder on black background with text Entry-level worker - refers to starting jobs ... More require little or no professional experience allow first jobbers just graduating to enter workforce There's a new report out on jobs this month, and it's not reassuring a lot of people who were worried about job displacement, especially for new graduates and less experienced workers. We know intuitively that that is a hard time to come of age as someone pondering a career. There are a lot of question marks around what AI is going to do to the job market. However, the wide-ranging report from Signalfire is showing us some early visions of what might come to pass in the years ahead. The scope of the study is impressive: it uses data from 650 million professionals, and 80 million organizations. There are breakout statistics for smaller groups of companies and workers, but the study in general relies on a pretty good survey of market data. One of the most sobering parts of the report is the idea that there's a generational shift taking place, and that it largely has to do with decreases in entry-level opportunities for new workers. For somebody just graduating from college, entry-level work is the name of the game. They can smooth the path with an internship or some kind of freelance position, but essentially, new career professionals have always faced that challenge of needing a job to get experience, and needing experience to get the job. However, what the study suggests is that this struggle has just gotten leaner and meaner – with a rise of 5.8% in unemployment for new college grads, and similar other harbingers of doom for Gen Z as a labor force. Other cited statistics include increases in certain law school admissions, with report writers theorizing that young professionals often put off job searches by attending law school when markets are tight. New graduate hires, the report shows, are down 50% from pre-pandemic levels. There's also the suggestion that non-technical job roles keep shrinking, and demand for high-tech positions (filled by experienced people) keeps growing, where senior people may be hired to fill junior positions. There's the question of whether college educations are keeping up with the skill sets employers are looking for. Fundamentally, the question that this asks is: if AI can do all of the go-fering, what happens to the go-fer jobs that have springboarded careers for decades? The study also analyzed where jobs are being created in the tech world. What they found is in increases in regional hubs like Miami and San Diego, lower hiring rates in parts of Texas, and continue dominance in San Francisco and New York where over 65% of the software engineers are located. The report explains it in this way, which to me seemed a little obtuse: 'More companies are embracing hub-and-spoke models, and tailoring compensation philosophies to ensure they secure the right talent mix across diverse locations.' The move toward larger hiring hubs makes sense in co-locating the talent with the infrastructure, and growing the economy of major cities, while emptying American small towns that weren't doing very well anyway. Economists and even public planners would presumably find it pretty easy to imagine how these trends will exacerbate the hollowing out of secondary or tertiary communities, and the building of massive metropolitan areas around hub cities. In other parts of the study, there's more analysis of how this could affect both workers and companies. I listen to an explication of the study from Nathaniel Whittemore at AI Daily Brief, where he went over a lot of the cultural aspects in play. He noted study suggestions that there may be other reasons for lower entry-level hiring, like higher interest rates and different company budget realities. But he also cited cultural business decisions, for example, in consulting and other areas, where some companies may just demand more from workers as a result of higher AI productivity. Whittemore cited engineers working for Amazon, where some of them have said that their job roles start to seem more like working in an Amazon warehouse, with the kinds of tight quotas and productivity mandates that you would associate with the word 'sweatshop'. 'I think that we're experiencing is an example of how challenging transitional periods can be,' he said, unpacking all of the uncertainties around how productivity gains will be handled. 'If you are a regular listener to this show, you'll know that I am net bullish on how all of this shakes out, I think that AI is going to bring massive disruption to the way that we work. ... I think that the market absorbs a huge amount of talent that would have otherwise been absorbed into these big companies in new and interesting ways. But that doesn't mean that it's not going to be extraordinarily painful along the way.' While companies, he notes, could train people better, they may just raise output goals, and be less forgiving of deadlines. Whittemore quotes a Harvard professor predicting a 'speed-up for knowledge workers,' which would represent the more partisan approach by employers – and unfortunately for new workers, that's not without precedent. There's a need, Whittemore suggests, for leadership, to determine whether there's a 'dehumanizing' or 're-humanizing' of human workers in the AI age. For that more targeted content from the report, here's an example, where in a study of the top 15 tech companies from 2019 to 2024, we see that there's almost a night and day difference on the chart lines between those with two or less years of experience, and others with more. It's scary for people who lack those first few years of career experience. In conclusion, I wanted to just post the last section of the report verbatim, because I think it speaks for itself: 'What it means for the road ahead: - For new grads: The training wheels are gone. With fewer entry-level roles, the path forward will rely on bootcamps, open-source, freelancing, and creative projects. It's not enough to just master the latest AI tools; learn to fix their flaws—debugging messy, machine-generated code may be your superpower. - For employers: AI might reduce the short-term need for junior hires, but skipping them entirely risks breaking the long-term talent pipeline. The industry's future depends on equipping the next generation with skills that grow alongside the evolving technology landscape.' So basically, doing too much in this area can hurt both the company, and the workers. Will we be able to find a solution that works for everybody? Anyway, this is a lot to take in for new hires or potential career professionals who are just graduating and entering the workforce. As Whittemore concedes, we are going to have to figure this out, and things could get messy before we end up with a solution for how to integrate AI changes into this aspect of our lives.
