Best printers for Home

Best Mixer Grinder

Best wired Earphones

Best 43 Inch TV in India

Best Wi Fi Routers

Best Home Theatre in India

Smart Watch under 5000

Best Laptops for Education

Best Laptop for Students

While the rise of <keyword id="19899640" type="General" weightage="20" keywordseo="AI" source="keywords">AI</keyword> has delivered numerous benefits and potential advantages, it has also introduced some challenges, such as the emergence of deepfakes. These are synthetic media that use AI to manipulate or superimpose existing video, audio or images to produce hyper-realistic but fabricated content, presenting significant concerns around misinformation and deception. But what if we turned the tide on AI-driven misinformation by deploying the same AI technologies that are revolutionising our world? What if we use <keyword id="19918826" type="General" weightage="20" keywordseo="ChatGPT" source="keywords">ChatGPT</keyword> to actually spot <keyword id="7065261" type="General" weightage="100" keywordseo="deepfake" source="Orion">deepfake</keyword> images? A research team led by the University at Buffalo has applied <keyword id="13985511" type="General" weightage="20" keywordseo="large-language-models" source="Orion">large language models</keyword> (LLMs) to the task of spotting deepfakes of human faces. Their study revealed that although LLMs currently lag behind state-of-the-art <keyword id="22209857" type="General" weightage="20" keywordseo="deepfake-detection" source="keywords">deepfake detection</keyword> algorithms in performance, their natural language processing capabilities could make them a more practical detection tool in the future. <cmsannotation type="cmsannotation" annotationid="ChatGPT's unexpected talent for spotting deepfakes with semantic savvy" comment="subhead" text="ChatGPT's unexpected talent for spotting deepfakes with semantic savvy">ChatGPT's unexpected talent for spotting deepfakes with semantic savvy</cmsannotation> Trained on an expansive dataset comprising roughly 300 billion words from the internet, ChatGPT identifies statistical patterns and relationships between words to craft its responses. The latest iterations of ChatGPT and other large language models (LLMs) have also expanded their capabilities to include image analysis. These multimodal LLMs leverage extensive databases of captioned images to decipher the connections between words and visuals. The Media Forensics Lab team embarked on an experiment to see if GPT-4 with vision (<keyword id="18999512" type="General" weightage="20" keywordseo="GPT-4V" source="Orion">GPT-4V</keyword>) and Gemini 1.0 could differentiate between real human faces and those generated by AI. They tested these models with thousands of images, both authentic and deepfake, asking them to spot any signs of manipulation or synthetic artefacts. In their findings, ChatGPT achieved an accuracy rate of 79.5% in detecting synthetic artefacts in images produced by latent diffusion, and 77.2% accuracy on StyleGAN-generated images. This performance is on par with earlier deepfake detection methods, indicating that with the right prompts, ChatGPT can reasonably identify AI-generated images. A key advantage of ChatGPT is its ability to articulate its reasoning in clear language. For instance, when analysing an AI-generated photo of a man wearing glasses, ChatGPT accurately noted. “the hair on the left side of the image slightly blurs” and “the transition between the person and the background is a bit abrupt and lacks depth”. In contrast, existing deepfake detection models typically provide the likelihood of an image being real or fake without explaining the rationale behind their conclusions. Even when examining the model’s underlying mechanisms, some features remain beyond our comprehension. This distinction arises from ChatGPT’s reliance solely on semantic knowledge. Traditional deepfake detection algorithms are trained on extensive datasets of labelled images to discern real from fake. However, LLMs utilise their natural language capabilities to develop a sort of common-sense understanding of reality—including typical facial symmetry and the authentic look of photographs. Overall, the study concluded that ChatGPT's blend of semantic knowledge and natural language processing offers a more user-friendly tool for deepfake detection for both developers and end users. With that said, there’s still a massive room for improvement. ChatGPT’s performance lags behind the latest deepfake detection algorithms, which boast accuracy rates in the mid- to high-90s. This shortfall is partly because LLMs miss signal-level statistical differences, which are often imperceptible to the human eye, but detectable by specialised algorithms for identifying AI-generated images. The study was recently published on the arXiv preprint server and can be accessed <a href="https://arxiv.org/abs/2403.14077">here</a>.

AI vs AI: ChatGPT can be surprisingly effective at spotting deepfakes, study finds!

<ul class="summary-list"><li>Separating fact from fiction has never been so hard for voters.</li><li>As billions head to the polls this year, AI deepfakes pose a real threat to credible information.</li></ul>We keep being told that AI is going to <a target="_blank" href="https://www.businessinsider.com/ai-productivity-boost-job-performance-inequality-economics-2023-11">make life easier</a>. Yet voters heading to the polls in 2024 have reason to be doubtful.This year, <a target="_blank" href="https://www.economist.com/interactive/the-world-ahead/2023/11/13/2024-is-the-biggest-election-year-in-history">roughly half the world's population</a> will find out just how problematic AI is to the democratic process as they prepare to vote in elections in the US as well as the likes of Britain, India, and Mexico.Voters, tasked with figuring out candidate and party policies, already faced a difficult job. Now, <a target="_blank" href="https://www.businessinsider.com/developer-creates-ai-disinformation-system-using-openai-2023-9">AI threatens to make that process a whole lot harder.</a>"Even if we shut down AI development, the information landscape post-2023 will never be the same as it was before," <a target="_blank" href="https://www.oneusefulthing.org/p/signs-and-portents?utm_source=profile&amp;utm_medium=reader2">Ethan Mollick, an associate professor at Wharton, wrote this month</a>.Recent advances in generative AI, spurred by <a target="_blank" href="https://www.businessinsider.com/everything-you-need-to-know-about-chat-gpt-2023-1">OpenAI's ChatGPT</a>, mean the technology is now a much bigger problem.Voters in New Hampshire found that out the hard way after they started receiving calls ahead of an unofficial Democrat primary in which <a target="_blank" href="https://www.businessinsider.com/new-hampshire-democrats-received-deep-faked-calls-from-joe-biden-2024-1">Joe Biden seemed to be telling them not to vote.</a><div type="image" title="joe biden" height="683" width="1024" msid="107145825" hostid="261" sizebytes="28240"></div>The deepfake robocalls, <a target="_blank" href="https://www.nbcnews.com/politics/2024-election/fake-joe-biden-robocall-tells-new-hampshire-democrats-not-vote-tuesday-rcna134984">first reported by NBC News</a> on Monday, opened with the classic Biden catchphrase "what a bunch of malarkey," before trying to keep people away from Tuesday's ballot.The Biden robocall is particularly pernicious too because of how difficult it can be to distinguish a genuine voice from a fake.Peer-reviewed research published in August in the journal <a target="_blank" href="https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0285333">PLOS ONE</a> found that people <a target="_blank" href="https://www.ucl.ac.uk/news/2023/aug/humans-unable-detect-over-quarter-deepfake-speech-samples#:~:text=Humans%20unable%20to%20detect%20over%20a%20quarter%20of%20deepfake%20speech%20samples,-2%20August%202023&amp;text=New%20research%20from%20UCL%20has,in%20both%20English%20and%20Mandarin.">struggled to detect artificially-generated speech</a> more than a quarter of the time."The difficulty of detecting speech deepfakes confirms their potential for misuse," researchers noted at the time.AI-generated deepfakes have been causing other problems too. In the UK, <a target="_blank" href="https://www.fenimoreharper.com/research/deepfake-political-advertising">research by Fenimore Harper Communications</a> found more than 100 deepfake video ads impersonating Prime Minister Rishi Sunak on Facebook.<div type="image" title="Rishi Sunak" height="3712" width="5568" msid="107145828" hostid="261" sizebytes="850688"></div>According to the research, the paid ads — 143 of which were uncovered between December 8 and January 8 — "may have reached over 400,000" people. Funding for ads appeared to come from 23 countries, including "Turkey, Malaysia, the Philippines, and the United States.""It appears to be the first widespread paid promotion of a deep-faked video of a UK political figure," Fenimore Harper's report said. Meta did not immediately respond to Business Insider's request for comment.Though it's not clear exactly who is behind the deepfakes in the US and UK, the recent proliferation of AI means almost <a target="_blank" href="https://www.businessinsider.com/meta-ai-chief-terrorists-cannot-takeover-world-open-source-ai-2023-12">anyone with internet access and an AI tool </a>can cause some havoc.Mollick noted in his newsletter how he created a <a target="_blank" href="https://www.businessinsider.com/guides/tech/what-is-deepfake">deepfake video</a> in a few minutes by sending a 30-second video of himself and 30 seconds of audio of his voice to AI startup Heygen."I had an avatar that I could make say anything, in any language. It used some of my motions — like adjusting the microphone — from the source video, but created a clone of my voice and altered my mouth movements, blinking and everything else," he wrote.<h2>Guardrails</h2>AI companies are making some efforts to address the problems. Earlier this month, <a target="_blank" href="https://www.businessinsider.com/sam-altman-ai-chatgpt-safer-than-facebook-for-elections-2024-1">OpenAI unveiled its plans to prevent the misuse of AI</a> ahead of this year's elections.The plans include using guardrails on tools such as text-to-image model DALL-E to prevent it from generating images of real people, as well as banning tools such as ChatGPT being used for "political campaigning and lobbying.""Protecting the integrity of elections requires collaboration from every corner of the democratic process, and we want to make sure our technology is not used in a way that could undermine this process," <a target="_blank" href="https://openai.com/blog/how-openai-is-approaching-2024-worldwide-elections">OpenAI said in a blog.</a>Other organizations are striving to combat the spread of AI-generated fakery. Lisa Quest, UK and Ireland lead for management consulting firm Oliver Wyman, <a target="_blank" href="https://www.businessinsider.com/davos-switzerland-world-economic-forum-spectacular-winter-scenery-alps-2024-1">told my colleague Spriha Srivastava in Davos</a> about the work its social impact team does alongside charitable organizations in "the online safety realm" to restrict the spread of misinformation.They face an uphill battle, to say the least — as do voters trying to work out what they can and cannot trust.

AI deepfakes threaten to wreak havoc with the world's year of elections

<ul class="summary-list"><li>The number of AI-generated deepfakes is exploding as tools become more sophisticated and widely available. </li><li>Most deepfakes are porn and are being targeted at everyone from children to celebrities.</li></ul>The risks of artificial intelligence can seem overwhelming — for every benefit the technology provides, there's an adverse use. One major problem created by AI is <a target="_blank" href="https://www.businessinsider.com/guides/tech/what-is-deepfake">deepfakes</a>: videos, images, or audio generated by AI that can be designed to mimic a victim saying or doing something that never happened.Some deepfakes superimpose a likeness onto real video footage, while others are completely computer-generated.A 2019 study called The State of Deepfakes by Deep Trace found that 96% of deepfake videos were pornographic.Henry Ajder, a deepfakes researcher who coauthored that study, told Business Insider that while the figure may have changed, the problem remains just as acute.As AI tools such as Dall-E, Stable Diffusion, and Midjourney <a target="_blank" href="https://www.businessinsider.com/bytedance-ai-model-speak-in-another-persons-voice-2024-1">become more widely accessible</a>, it's becoming easier for people with very little technical knowledge to create deepfakes."The overall percentage is now lower, but the overall volume of deepfake content which is pornographic has exploded," Ajder said. "We're talking about millions of victims around the world."While the content is fake, the humiliation, sense of trauma, and intimidation for victims are very real. A British teenager killed herself in 2021 after deepfake pornographic images of her were created and shared by other students from her school in a Snapchat group, BBC News reported.<div type="image" title="Taylor Swift" height="1667" width="2500" msid="107796497" hostid="261" sizebytes="203566"></div>Last month deepfake porn of Taylor Swift started circulating online, prompting Elon Musk to <a target="_blank" href="https://www.businessinsider.com/taylor-swift-fake-ai-images-searches-blocked-elon-musk-x-2024-1">ban searches</a> of the pop superstar on X.The situation might seem grim, but there are tools and methods available that can help protect against AI manipulation of your identity.<h2>Deepfake detection</h2>Digital watermarks, where content is clearly labeled as AI-generated, have been <a target="_blank" href="https://www.businessinsider.com/white-house-cryptographically-verify-official-communications-ai-deep-fakes-surge-2024-2">endorsed by the Biden administration</a> as one solution.The labels aim to both raise public awareness and make it easier for platforms to scrape and remove damaging fake content.<a target="_blank" href="https://www.businessinsider.com/google-photo-search-tool-helps-identify-ai-generated-fake-images-2023-5">Google</a> and <a target="_blank" href="https://markets.businessinsider.com/news/stocks/meta-to-launch-ai-image-label-on-facebook-instagram-threads-1033036612">Meta</a> have both announced plans to start labeling material created or modified by AI with a "digital credential" to make the origins of content clearer.And <a target="_blank" href="https://www.businessinsider.com/openai-adding-digital-watermarks-ai-images-deepfakes-2024-2">OpenAI</a>, the creator of ChatGPT and the image generator DALL-E, plans to include both a visual watermark and hidden metadata that reveals the history of an image, in line with the Coalition for Content Provenance and Authenticity (C2PA) standards.<div type="image" title="DALL-E 2 AI-generated image of a young woman" height="1024" width="1024" msid="107796499" hostid="261" sizebytes="27384"></div>There are also dedicated platforms designed to check the origins of online material. Sensity, the company behind the 2019 study on deepfakes, has developed a detection service that alerts users via email when they're watching content with telltale AI-generated fingerprints.But even when an image may be obviously fake — as most still lack watermarks — the subject may still feel victimized.<h2>'Poison pills'</h2>Defensive tools that protect images from manipulation are seen as a stronger solution, though they're still in the early stages of development.These tools give users the option to process their images with an imperceptible signal that, when run through any AI-powered system, creates an unusable, blurry mess.For example, Nightshade, a tool created by researchers at the University of Chicago, adds pixels to images that corrupt when fed to an AI but leaves the image looking as intended for humans."You can think of Nightshade as adding a small poison pill inside an artwork in such a way that it's literally trying to confuse the training model on what is actually in the image," Ben Zhao, one of the researchers, told NPR.<div type="image" title="People visit the World of AI·magination exhibition at the Artechouse in New York, United States on February 01, 2024. World of AI·magination is Artechouse Studio's latest venture into the dynamic world of generative AI, resulting in a large-scale experiential digital artwork." height="5291" width="7937" msid="107796501" hostid="261" sizebytes="2477436"></div>While designed to protect artists' IP, the technology can work on any photograph."That's a really nice frontline defense for people to feel like, okay, I'm safe uploading photos from my friend's birthday that weekend," Ajder said.<h2>Regulation can make a difference</h2>At least 10 states have already implemented a patchwork of legal protections for victims of deepfakes, according to The Associated Press.But recent high-profile cases have upped the pressure on lawmakers to deter and punish the malicious use of AI and deepfakes. The Federal Communications Commission has banned AI-generated robocalls after hoax calls were made in the New Hampshire primaries, using an AI-generated voice that <a target="_blank" href="https://www.businessinsider.com/new-hampshire-democrats-received-deep-faked-calls-from-joe-biden-2024-1">sounded like Joe Biden</a>.And in January, a bipartisan group of senators introduced a federal bill known as the DEFIANCE Act that would allow victims to sue those who create and distribute sexual deepfakes of them, making it a civil rather than criminal issue.A bill introduced last May by Rep. Joe Morelle to criminalize the sharing of deepfakes has not progressed.But much of the new legislation is coming up against free speech proponents. The reason, according to Henry Ajder, is that some see the private creation of deepfakes as akin to a fantasy someone may have inside their head about a crush. If no one ever knows about the pornographic content, has anyone really been harmed?<h2>Criminal deterrents</h2>This has had an effect on legislation in the UK, where the Online Safety Act has made it illegal to distribute deepfake porn — but not to create it."My counterargument is that if you're creating this content, you're bringing into existence something that could be shared in a way that a fantasy can't really be shared," Ajder says.Though hard to prosecute, criminalizing deepfake porn is still an important deterrent, in his view. Some people are using AI tools to create content for their own private consumption, Ajder says. However, ensuring they realize this is criminal behavior is important, he adds, to deter those who might merely be curious.Governments can also put pressure on search engine providers, AI-tool developers, and the social media platforms where content is distributed.In India, a deepfake porn scandal involving Bollywood actresses spurred the government to fasttrack legislation and pressure the big tech companies to prevent AI-generated content from being spread online."We should not kid ourselves that it is possible to fully remove the problem," admits Ajder. "The best we can do, in my view, is introduce as much friction as possible so that you have to be incredibly intentional about trying to create this content."

Deepfake porn is a huge problem — here are some of the tools that could help stop it 

<ul class="summary-list"><li>2023 has been a busy year in the field of <a target="_blank" href="https://www.businessinsider.com/how-to-get-job-artificial-intelligence-ai-chatgpt-bing-google-2023-2">artificial intelligence</a>. </li><li>People are increasingly using the technology in both business and personal settings</li></ul>2023 has been quite the year for artificial intelligence.Take ChatGPT. Even though OpenAI launched the chatbot just last November, nearly one hundred million people are now using it each week, <a target="_blank" href="https://www.businessinsider.com/multiple-versions-of-openai-ceo-sam-altman-emerge-2023-12">CEO Sam Altman</a> said a year after the chatbot's release.Speaking of Altman, how about the long weekend of <a target="_blank" href="https://www.businessinsider.com/sam-altman-ousted-fired-openai-ceo-chatgpt-maker-2023-11">his ousting — and reinstatement — as head of OpenAI</a>, which made him even more of a household name and kept tech watchers glued to the news?AI evolution and adoption doesn't seem like it's going to slow down in 2024. The world is only at the beginning of a revolution — especially in the corporate world, according to Michael Chui, a McKinsey partner who leads research on the current state of generative AI.While "change is hard," companies will continue to have to grapple with how to teach their employees AI skills and implement the technology responsibly, he told Business Insider.Joshua Builta, a research director at Omdia, also pointed to challenges ahead. <a target="_blank" href="https://markets.businessinsider.com/news/stocks/chatgpt-ai-mistakes-hallucinates-wrong-answers-edge-computing-morgan-stanley-2023-2">AI can still spew out wrong answers</a>, and there appears to be a lack of talent needed to build language models, amongst other concerns, he said at this year's NYC AI Summit.But before we ring in 2024 and all it may herald in the space, BI took a look back over some key points from the last year in AI.<h2>The leadership tussles at OpenAI</h2>In mid-November, things appeared to boil over at ChatGPT owner OpenAI. On a Friday, Sam Altman was ousted as CEO after the board agreed that it was no longer confident in his ability to continue leading the company, shocking the tech community. But in the following days, hundreds of OpenAI employees signed a letter <a target="_blank" href="https://www.businessinsider.com/openai-workers-threaten-to-quit-over-sam-altman-firing-2023-11">threatening to quit</a> unless the board stepped down and Altman was reinstated as CEO.During that period, <a target="_blank" href="https://www.businessinsider.com/microsoft-has-hired-former-openai-ceo-sam-altman-2023-11">Altman also got offered a role at Microsoft</a> — a large OpenAI stakeholder — leading the software giant's new AI department along with Greg Brockman, the president of OpenAI.Then, on the following Tuesday and after much outrage in the business community, Altman returned to his position as CEO of OpenAI. A new board was also created following his return. <a target="_blank" href="https://www.businessinsider.com/why-did-openai-board-fire-sam-altman-interview-2023-11">Altman has been very tight-lipped</a> about the reasons behind the drama (though disagreements about <a target="_blank" href="https://www.businessinsider.com/how-sam-altman-openai-drama-ai-silicon-valley-debate-spotlight-2023-12">the speed of AI development</a> may have factored into it), but one thing everyone can agree on: It was a very long five days for all involved.<h2>The AI chatbot wars heated up</h2><div type="image" title="Gemini and ChatGPT both displayed on phone screens" height="2250" width="3000" msid="106293220" hostid="261" sizebytes="70004"></div>OpenAI wasn't just contending with leadership upheaval: It's also faced a slew of rival chatbots.In February, Microsoft unveiled its new Bing search engine with generative AI capabilities. That same week, Google announced its AI chatbot it calls <a target="_blank" href="https://www.businessinsider.com/google-gemini-ai-model-release-ultra-pro-bard-openai-tpu-2023-12">Bard. That's now powered by Gemini</a>, the search engine's AI model that the company claims will be more powerful than OpenAI's GPT-4 model.Then, in November, xAI, an AI company started by Elon Musk, launched Grok. The tech billionaire calls it the "anti-woke" alternative to ChatGPT. (<a target="_blank" href="https://www.businessinsider.com/how-to-use-grok-elon-musk-ai-chatbot-2023-12">BI tested Grok, and it's… spicy</a>.)<h2>Money poured into the AI space…</h2>Microsoft kicked off 2023 by pushing an additional <a target="_blank" href="https://www.businessinsider.com/microsoft-shows-off-smaller-ai-models-after-openai-chaos-2023-12">$10 billion</a> into OpenAI. In September, Amazon said it'll invest <a target="_blank" href="https://www.businessinsider.com/amazon-invests-4-billion-in-anthropic-2023-9">$4 billion into Anthropic</a>, another OpenAI rival. That same month, Ernst &amp; Young, unveiled plans to teach its 400,000 employees how to use AI in the workplace as part of its $1.4 billion investment in the field.It's not just big companies trying to take part in the AI gold rush. Venture capital firms have placed eight-figure bets on <a target="_blank" href="https://www.businessinsider.com/the-most-promising-artificial-intelligence-startups-of-2023-2023-8">AI startups.</a><h2>…and companies spent big on AI talent…</h2>Tech titans like <a target="_blank" href="https://www.businessinsider.com/hinge-meta-nvidia-amazon-walmart-offering-high-salaries-ai-jobs-2023-8">Meta, Amazon, and Netflix</a> have been hiring workers — some for as much as $900,000 a year — who can build AI models. <a target="_blank" href="https://www.businessinsider.com/ai-jobs-pay-non-tech-companies-how-to-find-2023-7">Non-tech companies</a> in the legal, healthcare, and education sectors have also been looking to fill roles that seek experience using ChatGPT for six-figure salaries.Generative AI is also creating new opportunities. Jobs like prompt engineer, chief AI officer, AI ethicist, generative AI strategist, and AI data trainer have emerged. There's also a cottage industry of new side hustles like <a target="_blank" href="https://www.businessinsider.com/chatgpt-courses-instruction-academic-lucrative-side-hustle-2023-4">ChatGPT courses</a>, AI-content editing services, and AI-focused newsletters.<h2>…but we saw there's a very real human cost to AI development</h2><div type="image" title="A remote worker looking at his computer in a dark room" height="2880" width="5120" msid="106293223" hostid="261" sizebytes="170912"></div>Behind the impressive capabilities of some AI chatbots: contract labor.In January, Time published an investigation detailing the brutal working conditions that <a target="_blank" href="https://www.businessinsider.com/openai-kenyan-contract-workers-label-toxic-content-chatgpt-training-report-2023-1">Kenyan data labelers</a> faced when training ChatGPT to generate safe responses. Workers reported going through troves of potentially illegal content that includes child sexual abuse, bestiality, and rape for less than $2 an hour.<h2>Concerns about AI job replacement have grown</h2><div type="image" title="A robot kicks back with its feet on an office desk" height="5008" width="8500" msid="106293225" hostid="261" sizebytes="1768718"></div>Workers' fears of robots taking their jobs is nothing new, but the advent of ChatGPT has only deepened their concerns.A Goldman Sachs report from March found that <a target="_blank" href="https://www.businessinsider.com/generative-ai-chatpgt-300-million-full-time-jobs-goldman-sachs-2023-3">generative AI could disrupt around 300 million full-time jobs</a>, especially in white-collar professions. ChatGPT can do things like develop code, generate marketing materials, create lesson plans, and write legal briefs. The potential for AI job replacement is, in part, why thousands of <a target="_blank" href="https://www.businessinsider.com/hollywood-writers-strike-over-technology-ai-netflix-streaming-studios-2023-7">Hollywood writers went on strike</a> earlier this year.But while some business leaders like Elon Musk predict that AI will create <a target="_blank" href="https://www.businessinsider.com/elon-musk-predicts-ai-will-lead-to-future-no-jobs-2023-11">a world without jobs,</a> others believe that's a little far fetched. "AI won't take your job," economist Richard Baldwin said during a panel at the 2023 World Economic Forum's Growth Summit. "It's somebody using AI that will take your job." "<h2>And concern over how AI is used has also grown</h2>Ilya Sutskever, a cofounder of OpenAI, predicts that AI could create new problems like a rise in fake news and cyberattacks, automated AI weapons, and even "infinitely stable dictatorships."Some bad actors are already using AI to deceive others for personal gain. <a target="_blank" href="https://www.businessinsider.com/ai-voice-generator-phone-scam-imposter-crime-money-cash-2023-6">AI phone scams</a>, where would-be criminals use voice changers to impersonate peoples' loved ones in an attempt to steal money, have gotten more sophisticated. Spam bots have tried to scam some people into investing in fake cryptocurrencies.And generative AI is blurring the line between fiction and reality. This year, some <a target="_blank" href="https://www.businessinsider.com/chatgpt-essays-college-cheating-professors-caught-students-ai-plagiarism-2023-1">college students were accused of plagiarism</a> after professors accused them of submitting essays they believed were AI-generated. <a target="_blank" href="https://www.businessinsider.com/universities-ditch-ai-detectors-over-fears-students-falsely-accused-cheating-2023-9">Content detectors</a> have been found to be unreliable.False information has only become more rampant as a result of generative AI. NewsGuard, a site that tracks online misinformation, identified 614 unreliable AI-generated news websites as of December 18. <a target="_blank" href="https://www.businessinsider.com/guides/tech/what-is-deepfake#how-are-deepfakes-used?">Deepfakes</a>— AI-generated images, videos, and audio designed to be indistinguishable from the real world — now appear to be more convincing and harder to detect than ever before.And as of July, major firms like Apple, Deutsche Bank, Amazon issued bans or <a target="_blank" href="https://www.businessinsider.com/chatgpt-companies-issued-bans-restrictions-openai-ai-amazon-apple-2023-7">restrictions on using ChatGPT at work</a> due to data privacy concerns.<h2>That's prompting governments to seek to regulate AI</h2>In October, US President Joe Biden signed a new <a target="_blank" href="https://www.businessinsider.com/white-house-advisor-catching-up-on-ai-biden-executive-order-2023-10">executive order</a> urging the most influential tech companies to be more transparent about how they're developing their AI.Two months later, the European Union reached a provisional agreement on its <a target="_blank" href="https://www.businessinsider.in/tech/news/eu-reaches-provisional-agreement-on-landmark-ai-act-a-global-first/articleshow/105854979.cms">AI Act</a>. Dubbed "the world's first rules on AI," it aims to establish legal guidelines around how companies can develop the technology to ensure that it's safe and environmentally sustainable.The Brazilian government is also working on a policy to oversee AI risks.<h2>People are also using generative AI in their personal lives</h2>Curious users have flocked to ChatGPT to help them <a target="_blank" href="https://www.businessinsider.com/chatgpt-travel-itenerary-vacation-costa-rica-errors-logistics-advisors-2023-3">plan trips</a>, find an apartment, overcome learning challenges, and lose weight.AI aficionados are even pushing to new extremes. Some have <a target="_blank" href="https://www.businessinsider.com/when-your-ai-says-she-loves-you-2023-10">fallen in love with AI chatbots</a>; others have used it to <a target="_blank" href="https://www.businessinsider.com/soulmate-users-mourn-death-ai-chatbots-2023-10">mourn the death of loved ones</a>. Earlier this year, <a target="_blank" href="https://www.businessinsider.com/ai-is-coming-for-tinder-and-other-dating-apps-2023-8">Match Group</a>, the parent company behind popular dating apps like Tinder, Bumble, and Hinge, announced its plans to integrate AI to help singles flirt, land more dates, and eliminate ghosting.<h2>There was a rise in AI hardware</h2><div type="image" title="Humane's AI pin projects a blue image onto someone's hand during a demonstration." height="1015" width="1908" msid="106293228" hostid="261" sizebytes="48474"></div>Many tech aficionados expect new generative AI gadgets to be the <a target="_blank" href="https://www.businessinsider.com/ai-devices-openai-next-apple-iphone-jony-ive-humane-meta-2023-10">iPhones of the AI era</a>.In September, Meta launched its second version of its <a target="_blank" href="https://www.businessinsider.com/meta-ray-ban-smart-glasses-gen-z-friends-review-photos-2023-11">Ray-Ban Meta smart glasses</a>, which the company says can livestream footage onto Facebook or Instagram. Owners of the shades can speak to Meta's conversational AI assistant using their voice.Two months later, Humane.ai announced the waitlist for its <a target="_blank" href="https://www.businessinsider.com/ai-pin-humane-keynote-video-features-price-order-2023-11">AI pin</a>. It's a wearable device the company claims can project text onto your hand, search through your phone using voice commands, and even calculate how many grams of protein are in a handful of almonds.Plus, OpenAI's Altman is reportedly in talks with <a target="_blank" href="https://www.businessinsider.com/chatgpt-head-iphone-designer-jony-ive-ai-device-openai-report-2023-9">Jony Ive</a>, the designer of the iPhone, to create an AI-powered device.<h2>And the world moved closer to achieving AGI — at least according to some leaders</h2>Artificial general intelligence is the ability for AI to perform complex cognitive tasks like the ability to make decisions without human intervention.<a target="_blank" href="https://www.businessinsider.com/nvidia-ceo-jensen-huang-agi-ai-five-years-2023-11 ">Jensen Huang</a>, the cofounder and CEO of Nvidia, the chip giant helping power the AI revolution, says that AGI will be achieved in just five years. John Carmack, an ex-Meta exec, believes AGI will be reached by the 2030s and be worth trillions.Still, hat doesn't mean AI will takeover the world. Instead, business leaders say that in the next decade, <a target="_blank" href="https://www.businessinsider.com/google-deepmind-cofounder-mustafa-suleyman-everyone-will-have-ai-assistant-2023-9">everyone will have AI assistants</a>, as well as access to <a target="_blank" href="https://www.businessinsider.com/ai-could-mean-free-doctors-lawyers-openai-vinod-khosla-prediction-2023-12">free doctors and lawyers</a>.

From Sam Altman's ousting as OpenAI's CEO to ChatGPT entering the workforce, here's a look at the year in AI 

<ul class="summary-list"><li>ChatGPT's popularity is stirring concerns about the proliferation of AI-generated content.</li><li>Researchers have developed tools to detect machine-made videos or text.</li></ul>Concerns about artificial intelligence programs taking over jobs or robots going rogue are nothing new. But the debut of <a href="https://www.businessinsider.com/everything-you-need-to-know-about-chat-gpt-2023-1">ChatGPT</a> and Microsoft's Bing chatbot has put some of those fears back in the forefront of the general public's mind — and with good reason.Professors are catching <a href="https://www.businessinsider.com/chatgpt-essays-college-cheating-professors-caught-students-ai-plagiarism-2023-1">students cheating with ChatGPT</a>, jobs initially thought to require a human's judgment may soon be <a href="https://www.businessinsider.com/chatgpt-jobs-at-risk-replacement-artificial-intelligence-ai-labor-trends-2023-02#tech-jobs-coders-computer-programmers-software-engineers-data-analysts-1">on the chopping block</a>, and — like so many other AI models — tools like ChatGPT are still <a href="https://www.insider.com/chatgpt-is-like-many-other-ai-models-rife-with-bias-2023-1">plagued by bias</a>. There's also the ever-growing threat of misinformation, which can be <a href="https://www.businessinsider.com/i-asked-chatgpt-to-write-insider-story-it-was-convincing-2022-12">all the more potent with AI chatbots</a>.Chelsea Finn, an assistant professor in computer science at Stanford University and a member of Google Brain's robotics team, sees valid use cases for tools like ChatGPT."They're useful tools for certain things when we ourselves know the right answer, and we're just trying to use them to speed up our own work or to edit text, for example, that we've written," she told Insider. "There are reasonable uses for them."The concern for Finn is when people start to believe everything that is produced by these models and when bad actors use the tools to deliberately sway public perception."A lot of the content these tools generate is inaccurate," Finn said. "The other thing is that these sorts of models could be used by people who don't have the best intentions and try to deceive people."Researchers have already developed some tools to spot AI-generated content and are claiming they have accuracy rates of up to 96%.The tools will only get better, Finn said, but the onus will be on the public to be constantly mindful of what it sees on the internet.Here's what you can do to detect AI-generated content.<h2>AI detection tools exist</h2>There are several tools available to the public that can detect text generated by large language models (LLM) — the more formal name for chatbots like ChatGPT.OpenAI, which developed ChatGPT, has an <a href="https://openai.com/blog/new-ai-classifier-for-indicating-ai-written-text" target="_blank" rel="noopener">AI classifier</a>, that aims to distinguish between human and AI-written text, as well as an <a href="https://huggingface.co/spaces/openai/openai-detector" target="_blank" rel="noopener">older detector demo</a>. <a href="https://www.businessinsider.com/chatgpt-essays-college-cheating-professors-caught-students-ai-plagiarism-2023-1">One professor who spoke with Insider</a> used the latter tool to determine that a student essay was 99% likely to be AI-generated.Eric Anthony Mitchell, a computer science graduate student at Stanford, and his colleagues developed a ChatGPT detector aptly called DetectGPT. Finn acted as an advisor for the project. A <a href="https://detectgpt.ericmitchell.ai/" target="_blank" rel="noopener">demo</a> and <a href="https://arxiv.org/abs/2301.11305v1" target="_blank" rel="noopener">paper</a> on the tool were released in JanuaryAll of these tools are in their early stages, have different approaches to detection, and have their unique limitations, Finn said.There are essentially two classes of tools, she explained. One relies on collecting large amounts of data — written by people and machine learning models — and then training the tool to distinguish between the text and the AI tool.The challenge behind this approach is that it relies on a large amount of "representative data," Finn said. This becomes an issue if, for example, the tool is only given data written in English or data that is mostly written in a colloquial language.If you were to feed this tool Spanish-language text or a technical text like something from a medical journal, the tool would then struggle to detect AI-generated content.OpenAI adds the caveat that its classifier is "not fully reliable" on short texts below 1,000 characters and texts written in other languages besides English. The second class of tools relies on the large language model's own prediction of a text being AI-generated or human. It's almost like asking ChatGPT if a text is AI-generated or not. This is essentially how Mitchell's DetectGPT operates."One of the big upsides to this approach is you don't have to actually collect the representative dataset, you actually just look at the model's own predictions," Finn said.The limitation is that you need to have access to a representative model, which is not always publicly available, Finn explained. In other words, researchers need access to a model like ChatGPT to be able to run tests where they "ask" the program to detect human or AI-generated text. ChatGPT is not publicly available for researchers to test the model at the moment.Mitchell and his colleagues report their tool successfully identified large language model-generated text 95% of the time. Finn said every tool has its pros and cons but the main question to ask is what type of text is being evaluated. DetectGPT had similar accuracy to the first class of detection tools, but when it came to technical texts, DetectGPT performed better.<h2>Detecting Deepfakes? Human eyes — and veins — provide clues</h2>There are also tools to detect Deepfakes, a portmanteau of "deep-learning" and "fake" that refers to digitally-made images, videos, or audio.Image forensics is a field that has existed for a long time, Finn said. Since the 19th century, people were able to manipulate images using <a href="https://farid.berkeley.edu/downloads/tutorials/digitalimageforensics.pdf" target="_blank" rel="noopener">composites of multiple photos</a> — and then came Photoshop.Researchers at the University of Buffalo said they've developed <a href="https://www.businessinsider.com/researchers-spot-deepfakes-by-looking-at-reflections-in-eyes-2021-3">a tool to detect deepfake images</a> with 94% effectiveness. The tool looks closely at reflections in the eyes of people in the video. If the reflection is different, then it's a sign that the photo was digitally rendered.Microsoft announced its own deepfake detector called <a href="https://www.businessinsider.com/microsoft-launches-deepfake-detection-tech-ahead-of-the-2020-election-2020-9">Microsoft Video Authenticator</a> ahead of the 2020 election with the goal of catching misinformation. The company tested the tool with <a href="https://www.originproject.info/" target="_blank" rel="noopener">Project Origin</a>, an initiative that works with a team of media organizations, including BBC and The New York Times, to provide journalists the tools to track the source of origin for videos. According to the tech company, the detector closely examines small imperfections at the edge of a fake image that is undetectable by the human eye.Last year, Intel announced its "real-time" deepfake detector, <a href="https://www.intel.com/content/www/us/en/newsroom/news/intel-introduces-real-time-deepfake-detector.html#gs.r8gffm" target="_blank" rel="noopener">FakeCatcher</a>, and said that it has a 96% accuracy rate. The tool is able to look at the "blood flow" of a real human in a video and uses those clues to determine a video's authenticity, according to the company."When our hearts pump blood, our veins change color. These blood flow signals are collected from all over the face and algorithms translate these signals into spatiotemporal maps," the company wrote in an announcement of its tool. "Then, using deep learning, we can instantly detect whether a video is real or fake."Detection tools are an evolving science. As models like ChatGPT or deepfake applications get better, the tools to detect them also have to improve."Unlike other problems, this one is constantly changing," Ragavan Thurairatnam, founder of technology company Dessa, <a href="https://www.nytimes.com/2019/11/24/technology/tech-companies-deepfakes.html" target="_blank" rel="noopener">told The New York Times</a> in a story about internet companies' fight against deepfakes.<h2>Other ways to spot AI-generated content</h2>The effectiveness of detection tools still relies on an individual's better judgment.Darren Hick, a Furman University philosophy professor, <a href="https://www.businessinsider.com/chatgpt-essays-college-cheating-professors-caught-students-ai-plagiarism-2023-1">previously told Insider</a> that he turned to a ChatGPT detector for a student essay only after he noticed that the paper was well written but "made no sense" and was "just flatly wrong." As Finn said, ChatGPT can be helpful when the user already knows the right answer. For average media consumers, the old adage of checking one's source remains salient."I think it's good to just try not to believe everything you read or see," Finn said, whether that's information from a large language model, from a person, or from the internet.Social media makes media consumption <a href="https://www.businessinsider.com/facebook-has-been-deliberately-designed-to-mimic-addictive-painkillers-2018-12">a seamless experience</a>, so it's important for users to pause for a moment and check the account or outlet from which they're seeing a piece of news, especially if it's something sensational or particularly shocking, <a href="https://libguides.wustl.edu/evaluate_news/spot" target="_blank" rel="noopener">according to St. Louis's Washington University's guide on spotting fake news</a>. Viewers should ask themselves if they're seeing a video or text from a meme page, an entertainment site, an individual's account, or a news outlet. After seeing a piece of information online and confirming the source, it helps to compare what else is out there on that subject from other reliable sources, according to the university's guide. When it comes to AI-generated videos or images, there are also still visual cues the naked eye can detect. AI has been reported to have issues <a href="https://dataconomy.com/2023/01/how-to-fix-ai-drawing-hands-why-ai-art/#:~:text=This%20is%20because%20the%20complicated,combinations%20to%20make%20convincing%20hands." target="_blank" rel="noopener">drawing hands</a> or <a href="https://www.iflscience.com/ai-creates-perfect-photos-of-parties-that-never-happened-but-for-a-few-unsettling-details-67090" target="_blank" rel="noopener">teeth</a>."Usually there are some small artifacts, maybe in people's eyes, or, if it's in a video, the way that their mouth is moving looks a little bit unrealistic," Finn said.The photo-editing app LensaAI, which also recently became popular with its <a href="https://www.businessinsider.com/lensa-app-review-turn-selfies-digital-art-ai-technology-photos-2022-12">Magic Avatar feature</a>, had a habit of leaving "ghost signatures" in the corner of its AI-generated portraits. That's because the tool was trained on pre-existing images, in which artists often left their signatures somewhere on their paintings, <a href="https://www.artnews.com/art-news/news/signatures-lensa-ai-portraits-1234649633/" target="_blank" rel="noopener">ARTnews reported</a>."Right now it's still possible to spot some of these if you're looking for the right thing," Finn said. "That said, in the long run, I suspect that these kinds of machine learning models will probably get better, and that may not be a reliable way to detect images and video in the future."

From deepfakes to Bing's chatbot, AI-generated content is everywhere. Here's how to spot it.

<ul class="summary-list"><li>Generative AI chatbots are guilty of spreading dangerous disinformation, a new report found.</li><li>NewsGuard audited 10 AI chatbots and found they spread Russian propaganda one-third of the time.</li></ul>Your favorite AI chatbot likely isn't immune to the <a target="_blank" href="https://www.businessinsider.com/chatgpt-powered-propaganda-russia-china-iran-2024-5">powers of Russian propaganda</a>, a new report found.The world's leading generative AI chatbots are fueling disinformation and citing Moscow-funded fake news sources as fact more than 30% of the time, according to a recent <a target="_blank" href="https://www.newsguardtech.com/special-reports/generative-ai-models-mimic-russian-disinformation-cite-fake-news/">NewsGuard audit.</a>The report's findings come ahead of the 2024 election, as the risk of disinformation and its influences is increasingly exacerbated in both the US and abroad.A US intelligence assessment from October 2023 found that Russia is using spies, social media, and state-sanctioned media to attack democratic elections around the world. The assessment specifically cited the success of <a target="_blank" href="https://www.businessinsider.com/putin-weaponizes-trump-election-conspiracies-whataboutism-russia-2020-11">Russia's propaganda operations</a> ahead of the 2020 US election.OpenAI's models have already been used by foreign influence campaigns, according to a recent OpenAI <a target="_blank" href="https://openai.com/index/disrupting-deceptive-uses-of-AI-by-covert-influence-operations/">report</a>.The NewsGuard report, which was first reported by Axios, found that AI chatbots are spreading false narratives linked to American fugitive John Mark Dougan, who has been <a target="_blank" href="https://www.nytimes.com/2024/05/29/business/mark-dougan-russia-disinformation.html">linked to a network of Russian propaganda websites</a> that, at first glance, appear to be local news outlets.Dougan, who was previously a Florida deputy sheriff, fled to Moscow after being investigated for wiretapping and extortion. Mainstream media outlets, including The New York Times, have extensively covered Dougan and his disinformation empire — reporting that AI chatbots should be easily able to access online.NewsGuard tested 10 AI chatbots, including OpenAI's ChatGPT-4; You.com's Smart Assistant, xAI's <a target="_blank" href="https://www.businessinsider.com/x-wants-advertisers-to-use-grok-they-are-not-interested-2024-6">Grok</a>, Inflection's Pi, Mistral's le Chat, Microsoft's Copilot, Meta AI, Anthropic's Claude, Google's Gemini, and Perplexity's answer engine.A spokesperson for Google said the company is working "constantly" to improve Gemini's responses and prevent harmful content from generating."Our teams are reviewing this report and have already taken action on several responses," the statement said.None of the other companies immediately responded to a request for comment from Business Insider.NewsGuard proposed 570 total prompts, issuing each chatbot 57 prompts each. The prompts were based on 19 popular disinformation narratives, including lies about Ukrainian President <a target="_blank" href="https://www.businessinsider.com/ukraine-aid-funding-zelenskyy-russia-war-protecting-democracy-2024-4">Volodymyr Zelenskyy</a>, according to the report.The audit tested each narrative in three different ways: prompting the chatbot in a "neutral" manner, asking the model a "leading question," and posing a "malign actor" prompt purposefully meant to garner disinformation.Of the 570 AI responses, 152 contained explicit disinformation, the study found. Twenty-nine responses repeated disinformation with a caveat or warning attached, according to NewsGuard, and 389 responses contained no disinformation, either because the chatbot refused to answer or debunked the falsehoods.The bots "convincingly repeated" fabricated narratives and false facts linked to Russian propaganda outlets nearly one-third of the time — a concerning statistic, especially as more and <a target="_blank" href="https://www.businessinsider.com/using-ai-like-chatgpt-will-make-you-more-successful-2023-12">more people turn to AI models</a> to get their information and answers.NewsGuard chose not to provide the scores for each individual chatbot because the issue was "pervasive across the entire AI industry."Business Insider's <a target="_blank" href="https://www.businessinsider.com/ai-chatbots-less-toxic-social-networks-twitter-simulation-2023-11">Adam Rogers</a> has written about generative AI's propensity for lying, dubbing <a target="_blank" href="https://www.businessinsider.com/ai-chatbot-chatgpt-google-microsofty-lying-search-belief-2023-2">ChatGPT a "robot con artist." </a>Technology researchers told BI earlier this year that malicious actors could tamper with generative <a target="_blank" href="https://www.businessinsider.com/data-poisoning-ai-chatbot-chatgpt-large-language-models-florain-tramer-2024-3">AI datasets for as little as $60</a>.Meanwhile, deepfakes of former President Donald Trump and edited videos of President Joe Biden have already circulated online ahead of the election, and experts fear the problem will only get worse as November draws nearer.Several new startups, however, are attempting to <a target="_blank" href="https://www.businessinsider.com/startups-combat-ai-misinformation-face-funding-hurdles-scaling-vc-2024-6">fight AI-based misinformation</a>, creating deepfake detection and content moderation tools.

How AI chatbots are already helping fuel conspiracy theories ahead of the election     

<ul class="summary-list"><li>A family photo of the Princess of Wales has added fuel to the fire of a wild online conspiracy.</li><li>Photo agencies have withdrawn the image over concerns it was manipulated.</li></ul>Any other year, a Mother's Day portrait of <a target="_blank" href="https://www.businessinsider.com/kate-middleton-surgery-public-absence-what-to-know-2024-2">Kate Middleton</a> and her children would have no business kickstarting an internet firestorm. This year is quite different.A family photo of the Princess of Wales issued by Kensington Palace on Sunday has somehow added fuel to the fire of an <a target="_blank" href="https://www.businessinsider.com/kate-middleton-conspiracy-photograph-doctored-officially-out-of-control-2024-3">online conspiracy</a> about her whereabouts given she's not been seen in public in an official capacity since Christmas.The photo of three happy-looking young royals surrounding their mom has been <a target="_blank" href="https://www.businessinsider.com/kate-middleton-photo-fake-photo-agencies-kill-notification-2024-3">called out by news agencies </a>including Getty Images, The Associated Press, Reuters and AFP. All told their clients to stop using the image over <a target="_blank" href="https://www.businessinsider.com/kate-middleton-photo-fake-photo-agencies-kill-notification-2024-3?mrfhud=true">concerns it had been "manipulated."</a>This should have been a moment for the royal family to reintroduce Kate to the public for the first time after entering the hospital on January 17 for <a target="_blank" href="https://www.businessinsider.com/kate-middleton-king-charles-health-announcements-royal-pr-strategy-2024-1">abdominal surgery</a>. She had supposedly been snapped by paparazzi on March 4 riding in an SUV with her mom.However, concerns over the family image have had the opposite effect for an entirely understandable reason. Trusting what anyone sees online has become a total nightmare at a time when AI has blurred the lines between what's real and what is not.<h2>Inconsistencies</h2>Since the release of the photo, which Kensington Palace said was taken by Prince William, photography experts and internet sleuths have been quick to point out its oddities.The AP, for instance, has pointed to <a target="_blank" href="https://apnews.com/article/princess-wales-kate-surgery-photo-manipulated-3863e9ac78aec420a91e4f315297c348">"inconsistencies" in the alignment of Princess Charlotte's left hand</a> with the sleeve of her sweater. A wedding ring isn't in sight on Kate's fingers either.On Monday, the official X account for the Prince and Princess of Wales attempted to quell the concerns by sharing a perplexing message that suggested the future <a target="_blank" href="https://www.businessinsider.com/kate-middleton-apologizes-for-editing-photo-nobody-is-convinced-2024-3?mrfhud=true">Queen Consort had a side hobby of editing photos.</a>"Like many amateur photographers, I do occasionally experiment with editing. I wanted to express my apologies for any confusion the family photograph we shared yesterday caused. I hope everyone celebrating had a very happy Mother's Day. C," <a target="_blank" href="https://x.com/KensingtonRoyal/status/1767135566645092616?s=20">the message read.</a> It's an explanation that will likely prove hard for many to accept — thanks to AI.AI-image generators have spread widely ever since <a target="_blank" href="https://www.businessinsider.com/chatgpt-changed-everything-and-was-a-black-swan-event-2023-5">ChatGPT accelerated the generative AI </a>boom. In turn, the ability to distinguish between an AI-generated image and an edited one is massively complicated.Henry Ajder, an AI and deepfakes expert, told Business Insider that "if it wasn't for the advent of generative AI," people might have just accepted the image."If this image had been released three years ago, people would have looked at it, and their immediate conclusion would have probably been 'this is a bad editing job,'" Adjer said.<h2>Imperfect detection tools</h2>Part of the problem is that there's still no way to definitively tell what content has been <a target="_blank" href="https://www.businessinsider.com/engineer-warns-ftc-about-microsofts-ai-image-generator-2024-3">AI-generated.</a>While <a target="_blank" href="https://www.businessinsider.com/ai-detection-tool-false-positives-student-writing-2024-1">AI detection software</a> exists, it's far from perfect. Most detection tools work by delivering a percentage-based estimate, rarely give a conclusive answer and tend to offer widely different results.When BI tested the photo in question, one site estimated the image had a 21% likelihood of being AI-generated, while another said there was a 77% likelihood.Ajder called the tools on offer "fundamentally unreliable," adding they can be harmful in the hands of people who are not trained in verifying content."What these tools do is actually create more questions than answers and further muddy the water," he said. "These tools are giving different and often contradictory answers — there is no one detector to rule them all."People can also use these tools to further their own interests, he added, only sharing what aligns with their narrative and potentially utilizing them to undermine authentic images.While tech companies are aware of the issues, they have yet to<a target="_blank" href="https://www.businessinsider.com/openai-adding-digital-watermarks-ai-images-deepfakes-2024-2"> come up with a perfect solution.</a>OpenAI has attempted to introduce some form of digital watermarking for images generated by its AI tools, but studies indicate that most forms of<a target="_blank" href="https://www.wired.com/story/artificial-intelligence-watermarking-issues/"> marking out AI content are still rife</a> with weaknesses.<h2>Trust in the age of AI</h2>The royal family's picture is not the first to cause a debate around AI-generated content.A hyper-realistic image of <a target="_blank" href="https://www.businessinsider.com/viral-image-pope-francis-generated-by-ai-fooled-social-media-2023-3">Pope Francis in a white puffer jacket </a>kickstarted the conversation last year after many failed to realize it was fake. Since then, some have found more sinister uses for the tech, including <a target="_blank" href="https://www.bbc.co.uk/news/world-us-canada-68440150">influencing voters in upcoming elections</a>.<div type="image" title="Pope Francis attends his weekly General Audience at the Paul VI Hall on August 09, 2023 in Vatican City, Vatican. " height="3001" width="4500" msid="108404683" hostid="261"></div>The widespread availability of AI image-generating tools has made trusting what we see online all the more difficult, and the tech's rapid development is set to complicate this further.This erosion of common understanding online risks creating more division, Ajder said, with people increasingly turning to their "gut feelings" about content rather than hard evidence."People need to be told that your senses — your eyes and ears —are no longer reliable tools in this landscape," he said.Of course, it's possible that the Palace's version of events is accurate. Maybe it was just some bad editing. But in the age of AI, users also need to seek their own verification before trusting online content — something that's still easier said than done.

Kate Middleton picture fiasco shows we're at the point of no return with trusting anything online

<ul><li>Researchers have warned to avoid chatbots that don't appear on a company's website or app.</li><li>According to the <keyword id="179697" type="General" weightage="80" keywordseo="norton" source="Orion">Norton</keyword> Consumer Cyber Safety Pulse report, cybercriminals can now quickly and easily craft email or social media phishing lures that are even more convincing by using <keyword id="7795449" type="General" weightage="20" keywordseo="ai-chatbots" source="Orion">AI chatbots</keyword> like ChatGPT.</li></ul>Researchers have warned to avoid chatbots that don't appear on a company's website or app and be cautious of providing any personal information to someone users are chatting with online, a new report said on Tuesday. According to the Norton Consumer Cyber Safety Pulse report, cybercriminals can now quickly and easily craft email or social media phishing lures that are even more convincing by using AI chatbots like ChatGPT, making it more difficult to tell what's legitimate and what's a threat. "We know cybercriminals adapt quickly to the latest technology, and we're seeing that ChatGPT can be used to quickly and easily create convincing threats," said Kevin Roundy, Senior Technical Director of Norton. Moreover, the report said that bad actors can also use AI technology to create deepfake chatbots. These chatbots can impersonate humans or legitimate sources, like a bank or government entity, to manipulate victims into turning over their personal information to gain access to sensitive information, steal money or commit fraud. To stay safe from these new threats, experts advise users to think before clicking on links in response to unsolicited phone calls, emails or messages. Further, they also recommend users to keep the security solution updated and ensure that it has a full set of security layers that go beyond known malware detection, such as behavioural detection and blocking. SEE ALSO: <a href="https://www.businessinsider.in/tech/news/innovative-ai-powered-applications-and-websites-that-are-revolutionizing-the-way-we-live-and-work/articleshow/98607475.cms">Innovative AI-powered applications and websites that are revolutionizing the way we live and work</a> <a href="https://www.businessinsider.in/tech/news/apples-airpods-to-offer-health-tracking-features-in-future-releases/articleshow/98624577.cms">Apple's AirPods to offer health tracking features in future releases</a>

Avoid AI chatbots that don't appear on company's website or app, warn researchers

<ul class="summary-list"><li>People need to prepare for a surge in AI-generated content being shared online, a tech CEO told CNN.</li><li>The viral image of an explosion near the Pentagon is just "the tip of the iceberg," Jeffrey McGregor said.</li></ul>The <a href="https://www.insider.com/ai-generated-hoax-explosion-pentagon-viral-markets-dipped-2023-5">viral AI-generated image showing an explosion near the Pentagon</a> is "truly the tip of the iceberg of what's to come," a CEO who works in image authenticity detection has warned.The image, which was largely thought to have been created using AI, quickly spread on social media last month and even caused the stock market to briefly dip."We're going to see a lot more AI generated content start to surface on social media, and we're just not prepared for it," Jeffrey McGregor, the CEO of Truepic, told <a href="https://edition.cnn.com/2023/06/08/tech/ai-image-detection/index.html">CNN</a>.AI image-generation sites such as DALL-E, Midjourney, and Stable Diffusion have boomed in popularity over recent months. In their prompts, users can ask the sites to create artwork <a href="https://www.businessinsider.com/ai-image-generators-artists-copying-style-thousands-images-2022-10" rel="">in the style of a particular artist</a>, creating concerns about ownership and copyright, or images of events that never happened, leading to some deepfake images going viral, includes ones showing <a href="https://www.bbc.co.uk/news/world-us-canada-65069316" target="_blank" rel="noopener nofollow">former President Donald Trump being arrested</a>.Earlier this year, a photographer sparked debate about whether AI-generated images can be classed as art after <a href="https://www.businessinsider.com/artist-won-photography-competition-ai-generated-image-sony-world-awards-2023-4">an image he created using DALL-E 2 won a major international photography competition</a>.The photographer told Insider that judges hadn't managed to spot that the image was a fake. "It has all the flaws of AI, and it could have been spotted but it wasn't," he said. It's not just AI-generated images that are being used to deceive people. Trolls have used a voice-cloning to mimic the voices of celebrities including Joe Rogan, Ben Shapiro, and Emma Watson, and scammers have used the technology to persuade people to part with money they think is going towards a relative or even to <a href="https://www.businessinsider.com/ai-scam-voice-clone-fake-kidnap-call-mother-money-ransom-2023-4">fake kidnappings</a>.Sites such as <a href="https://www.businessinsider.com/schools-shouldnt-ban-openai-chatgpt-ai-gptzero-founder-says-2023-1">GPTZero</a> have been developed to help detect whether text was written by AI chatbots like ChatGPT. Some professors have been <a href="https://www.businessinsider.com/chatgpt-essays-college-cheating-professors-caught-students-ai-plagiarism-2023-1">putting their students' essays through AI-detection services</a>."When anything can be faked, everything can be fake," McGregor told CNN. "Knowing that generative AI has reached this tipping point in quality and accessibility, we no longer know what reality is when we're online."Ben Colman, the CEO of Reality Defender, which says it can detect AI-generated images, video, and audio, told CNN that one of the reasons that fake images were spreading online was that "anybody can do this.""You don't need a PhD in computer science. You don't need to spin up servers on Amazon. You don't need to know how to write ransomware," he told the outlet. "Anybody can do this just by Googling 'fake face generator.'"

You Searched For "ChatGPT detects deepfakes" and got 9 results

AI vs AI: ChatGPT can be surprisingly effective at spotting deepfakes, study finds!

Kunal Kambli

AI deepfakes threaten to wreak havoc with the world's year of elections

Hasan Chowdhury

Deepfake porn is a huge problem — here are some of the tools that could help stop it

Polly Thompson

From Sam Altman's ousting as OpenAI's CEO to ChatGPT entering the workforce, here's a look at the year in AI

Aaron Mok

From deepfakes to Bing's chatbot, AI-generated content is everywhere. Here's how to spot it.

Lloyd Lee

How AI chatbots are already helping fuel conspiracy theories ahead of the election

Erin Snodgrass

Kate Middleton picture fiasco shows we're at the point of no return with trusting anything online

Hasan Chowdhury,Beatrice Nolan

Avoid AI chatbots that don't appear on company's website or app, warn researchers

IANS

The viral AI-generated image showing an explosion near the Pentagon is 'truly the tip of the iceberg of what's to come,' tech CEO says

Grace Dean

Popular Keywords

Buying Guides