Bunnings has switched on facial recognition technology at two of its stores. It marks the start of a phased nationwide rollout, with the hardware brand saying it will be used to protect staff and customers and to reduce theft by repeat offenders amid an increase in "threatening incidents". The first two stores to use the technology were Bunnings Te Rapa and Hamilton South. The technology would eventually be rolled out nationwide. As the rollout began, Bunnings shared a series of CCTV videos showing violent incidents at its stores across the country. One, taken from a Bunnings in Porirua, shows a man approaching staff with a knife as he pushes two trolleys full of items out of the store. Another, taken in Takanini, shows a person smashing into a Bunnings worker with a large box before running away. Meanwhile, another person picks up one of the boxes. Footage from New Lynn shows a man following a worker, waving his hands. When another staff member steps in, the man charges at the employees with a basket, bowling them over. Two more workers approached the scene and were also forced away by the man. A video from Bunnings Palmerston North showed a group of customers trying to stop a person pushing a trolley out of the store. When someone pulls them away, a brief scuffle breaks out on the floor before they flee out the door. Footage from the carpark of Bunnings Takanini showed a person being punched after getting out of their car by a hooded man and being chased away. Bunnings New Zealand general manager Melissa Haines said the technology was a "tool that helps us address the rising problem of violent, threatening and abusive behaviour and high-value theft, identify serious repeat offenders and act early to prevent incidents". “Threatening incidents
Apr 14, 2026 · via 1news.co.nz
‘That wasn’t me’: How facial recognition led to a woman being jailed for 6 months HERMITAGE, Pa. - To investigators, the case against Kimberlee Williams appeared straightforward. “It’s very obvious it’s you,” an officer in Montgomery County, Maryland, said to Williams, who was handcuffed to a table in the police department. Williams had several prior convictions for writing bad checks in Oklahoma, and the person who appeared in surveillance photos, accused of making fraudulent withdrawals in Maryland, looked like a match in surveillance photos - the same build, the same age, the same shape face. But Williams was adamant. That wasn’t her. She lived in Oklahoma and had never been to Maryland, she said, until police arrested her and flew her in to be questioned. Williams, now 57, offered to take a polygraph test and said she had family members who could provide an alibi for her, according to video of the interview from July 2021 reviewed by The Washington Post. She acknowledged that she had a history of writing bad checks, but she insisted that was in the past. “I’m not trying to waste your time,” Williams said. “I’m telling you -” “You’re telling me, but you’re not telling me the truth,” the officer interrupted. Williams was charged in three Maryland counties - Montgomery, Prince George’s and Anne Arundel - and jailed for six months before the charges were dismissed, according to court records. She would not learn for years why she how she came to be accused: A bank investigator reported to Montgomery County police that facial recognition technology had identified Williams as the suspect in the Maryland fraud cases. Montgomery County police did not disclose that Williams had been identified using facial recognition when they sought charges against her, according to police records. Several criminal justice experts
Apr 14, 2026 · via spokesman.com
More than a Dozen Wrongful Arrests Due to Police Reliance on Facial Recognition Technology When police arrested Kimberlee Williams, a grandmother living in Oklahoma, because of a warrant from Maryland, she was shocked. She had never been to Maryland in her life. Ms. Williams later learned that Maryland police had relied on an incorrect result from facial recognition technology that falsely flagged her as a suspect. She is the fourteenth person in the U.S. to join a growing list of people wrongfully arrested because police let flawed facial recognition technology taint their investigations. Police use of facial recognition technology is dangerous, and stories of people wrongfully arrested because of police reliance on incorrect facial recognition results continue to surface. Today, the ACLU and ACLU of Maryland sent letters to three Maryland police departments on behalf of Ms. Williams, who was wrongfully arrested and jailed for six months because Maryland police relied on a false facial recognition result and concealed their reliance on that unreliable technology from the court when applying for an arrest warrant. One Woman Arrested for a Crime She Didn't Commit On June 23, 2021, Ms. Williams was accompanying one of her daughters on a DoorDash delivery to a local military base in Lawton, Oklahoma. When base security at the entry checkpoint conducted a standard identification check, they discovered outstanding Maryland arrest warrants for Ms. Williams and detained her. These warrants sought Ms. Williams’ arrest for a series of fraudulent over-the-counter cash withdrawals in Maryland in December 2019 and January 2020. An unknown individual had entered SunTrust and Truist bank branches in three different counties, impersonated account holders, and fraudulently withdrew thousands of dollars from those individuals’ accounts. Ms. Williams, however, was nowhere near Maryland during this time. She was a resident of Oklahoma, living with two of
Apr 14, 2026 · via aclu.org
A study on vector database and AI integration identifies unstable indexing, weak cross-modal fusion, and rigid resource scheduling as key barriers. By introducing HNSW optimization, unified feature alignment, and dynamic computing allocation, the research improves retrieval accuracy, system responsiveness, and scalability for large-scale AI infrastructure. New York, NY, United States, April 14, 2026 — Vector databases have become a foundational component of modern AI infrastructure, enabling the high-dimensional search and semantic retrieval capabilities that power applications ranging from question answering to image recognition. The study Cutting-Edge Challenges and Solutions for the Integration of Vector Database and AI Technology examines critical barriers in vector database and AI integration, identifying unstable vector index accuracy, inefficient cross-modal feature fusion, and rigid computing resource scheduling as core obstacles. The research proposes targeted solutions using HNSW graph architectures, a unified feature fusion framework, and dynamic resource scheduling to improve retrieval accuracy, response speed, and scalability across production AI systems. Vector database systems face persistent technical challenges that limit the reliability and performance of modern AI applications, including indexing instability under high-dimensional search conditions, misalignment between different data modalities, and inflexible resource allocation models that cannot adapt to variable workloads. The research identifies where current vector database infrastructure falls short and proposes concrete solutions to improve system performance at scale, with applications spanning question answering, image recognition, and multilingual dialogue systems where retrieval accuracy and low latency are critical to reliable operation. The study puts forward three targeted solutions corresponding to each identified challenge. To stabilize vector index performance, the research proposes search structure optimization using Hierarchical Navigable Small World (HNSW) graph architectures, which allow systems to locate similar data points more reliably without scanning entire datasets. To address cross-modal feature fusion, the study introduces a unified integration framework that brings different data representations into semantic
Apr 14, 2026 · via usatoday.com
About
Press
Copyright
Contact us
Creators
Advertise
Developers
Terms
Privacy
Policy & Safety
How YouTube works
Test new features
NFL Sunday Ticket
© 2026 Google LLC
Apr 14, 2026 · via youtube.com
Woman Wrongly Jailed for Months Based on Faulty Facial Recognition Technology Demands Apology from Maryland Police Departments In the letters sent by the ACLU, Kimberlee Williams also lays out the reforms needed to help ensure this never happens to someone else MARYLAND — Today, the American Civil Liberties Union and ACLU of Maryland sent letters to three Maryland police departments on behalf of Kimberlee Williams, an Oklahoma woman who was wrongfully arrested because Maryland police relied on an incorrect result from facial recognition technology and concealed their reliance on that unreliable technology from the court when applying for arrest warrants. Ms. Williams is the fourteenth person publicly known to have been wrongfully arrested by U.S. police because of reliance on erroneous facial recognition results. “I lost six months of my life when Maryland police wrongfully imprisoned me halfway across the country from my children, my home, and my job, all because they relied on an incorrect result from faulty technology,” said Kimberlee Williams. “I had never even been to Maryland before I was flown there in handcuffs, for a crime I had nothing to do with. My family and I can’t get that time back, but I hope my experience will be a warning to police in Maryland and across the country that this technology can ruin lives. No family deserves to go through that.” On June 23, 2021, Ms. Williams was accompanying one of her daughters, a DoorDash driver, as she made a delivery to a local military base in Lawton, Oklahoma. When base security conducted a standard ID check, they discovered an outstanding Maryland arrest warrant for Ms. Williams, detained her, and called local police, who arrested her. Ms. Williams spent 23 days in an Oklahoma jail before a Maryland officer arrived to transport her to a jail
Apr 14, 2026 · via aclu.org
Apple is testing four smart glasses designs to compete with Meta, report says Many Silicon Valley heavy hitters are investing heavily in smart glasses, and a new report sheds light on Apple's plans to join the party. The latest report from Bloomberg's Mark Gurman claims that Apple is currently testing four different designs for smart glasses, which would compete with the Meta Ray-Ban Display glasses. Gurman says the product could be unveiled in late 2026 ahead of a 2027 launch. A lot could change in that time, though Gurman has a very impressive track record of Apple scoops. You May Also Like The Bloomberg report states that Apple is testing a variety of frame styles, such as rectangular, oval, and circular, as well as multiple color options. Some or all of these designs could be available upon launch if Apple smart glasses do come to market. In addition, the glasses would feature unique oval camera lenses, oriented vertically, with lights, according to Bloomberg. Finally, onboard speakers would allow users to take phone calls and listen to music, similar to other AR glasses from companies like Xreal and TCL. Based on previous reports, Apple will be making augmented reality glasses with a built-in display, similar to Meta Ray-Ban Display smart glasses and the unreleased Google Android XR glasses expected to be released later this year. Presumably, by the time Apple smart glasses are released, Apple will have also introduced the long-awaited Siri AI upgrade. That could allow users to talk to Siri while wearing the glasses for voice control, help with navigation, or activate other AR features. Meta-Ray Ban smart glasses sales tripled in 2025, per Reuters, but smart glasses have not yet gone mainstream the way smartphones, tablets, or laptops have. Big Tech companies see this new form factor as
Apr 14, 2026 · via mashable.com
In a matter of days, MAGA media figures have gone from defending President Donald Trump as God’s chosen president to making the case that he is actually the antichrist. The accusations reached a fever pitch on Sunday night, after Trump posted an AI-generated image depicting himself dressed in a white robe and red sash and placing a hand radiating golden light on a man in a hospital bed. The image, which Trump deleted from Truth Social after around 14 hours, seemed to invoke Jesus Christ, and it outraged some supporters who likened Trump’s behavior to the antichrist—a figure in Christian theology who opposes Christ and whose appearance many believe could augur the end of time. Major figures in the MAGA universe quickly spoke out. “It’s more than blasphemy. It’s an Antichrist spirit,” former congressperson Marjorie Taylor Greene posted to X on Sunday. “In 18 months I went from hesitantly voting for Trump to thinking there’s a decent chance he’s the antichrist,” added Clint Russell, host of the right-wing Liberty Lockdown podcast. “I genuinely believe Trump is currently demon possessed,” far-right Texas pastor Joel Webbon wrote on X. Hours later, Webbon hosted a livestream chat titled “Is Donald Trump the Anti-Christ?” The Knights Templar Order, a Christian organization based on a medieval military order, wrote of the post that they had “no other choice but to condemn it wholeheartedly and ask for a public apology.” For more than a decade, Trump and his supporters have used explicitly religious rhetoric and images to mobilize his base. But in recent weeks, some of Trump’s actions—from posts about the Vatican to messages about Iran posted on Easter Sunday—have caused a major schism among his supporters that could have lasting effects for him and the Republican Party. The White House did not respond to a
Apr 13, 2026 · via wired.com
This transcript has been edited for clarity. Welcome to Impact Factor, your weekly dose of commentary on a new medical study. I’m Dr F. Perry Wilson from the Yale School of Medicine. There’s a new genre of medical papers in the “AI in medicine” space, and, like Mulder from The X-Files, I want to believe. The theme of these papers is something like “LLMs aren’t going to replace good doctors,” and that is very reassuring for this doctor, who definitely does not want to be replaced by a friendly AI. But if I’m honest with myself, my belief in human supremacy here is being shaken. This week, a new paper appeared in JAMA Network Open evaluating the performance of various large language models on a diagnostic task. It was chock full of phrases like “our evaluation suggests that despite rapid advances in pattern recognition and knowledge retrieval, current LLMs still lack the reasoning processes needed for safe clinical use.” That’s reassuring. It also states that “the promise of LLMs in clinical medicine lies in their potential to augment — not replace — physician reasoning.” Perfect. I’m happy to be augmented. I would rather not be replaced. But then I read through the study and, frankly, I’m more worried now than ever. Let me break it down for you. Researchers evaluated 21 off-the-shelf large language models, including all the major players (ChatGPT, Claude, DeepSeek, Grok, Gemini), across 29 clinical vignettes from the MSD manual. The key innovation of this study over previous evaluations is how these vignettes are structured. Other studies evaluating LLMs often present an entire case and simply ask “what’s the diagnosis?”, but that’s not how these clinical vignettes work. Instead, they develop iteratively. You get an initial presentation and then formulate a differential diagnosis, choose the appropriate tests,
Apr 13, 2026 · via medscape.com
The trial by British immigration authorities of live facial recognition technology at Holyhead represents a significant step towards mass surveillance. The limited use of facial recognition for automated immigration control – the traveller is matched with a passport database – has been a feature of airports both here and abroad for over a decade . What is different in this case is that the technology is being used to match travellers to unspecified “watchlists” in real time. In the case of the Holyhead trial the target was reported to be people who had been removed from the UK and were seeking to re-enter. There is little that civil liberties campaigners here and in the UK can do about the trials and the clear direction of travel. The legality of the technology in the UK is well established and it has been used sporadically by British police forces since 2020. But the results of the trial – which have been published – provide useful ammunition to challenge the introduction of live facial recognition here. The Garda Síochána (Recording Devices) (Amendment) Bill 2025 is making its way through the Dáil. It will allow the use of automated recognition technology and also live recognition technology. The background is the 2023 Dublin riots, when the Garda had to manually review hours of CCTV footage to identify offenders. READ MORE The Irish Council for Civil Liberties and others have argued that the UK trial cast doubts over the efficiency and efficacy of the routine deployment of live facial recognition. Over 10,000 passengers at Holyhead were screened against a watchlist of up to 6,535 Individuals, generating two alerts. There is a limit to the conclusion that can be drawn from the limited trial, but the ICCL is right to question whether the technology is necessary or
Apr 13, 2026 · via irishtimes.com
As Meta continues to advance its plan to make artificial intelligence-powered glasses a key factor for digital connection, a group of more than 70 advocacy organizations have issued a warning about the potential invasion of privacy that these devices could facilitate. The alarm for regulators comes ahead of a broader launch of Meta’s latest update. As reported by Wired, a coalition of more than 70 civil liberties, domestic violence, reproductive rights, LGBTQ+, labor and immigrant advocacy organizations issued a demand that Meta abandon its plans to deploy face recognition in its AI glasses, due to concerns that this could enable stalkers, abusers and federal agents to covertly identify strangers in public. In February, a report from the New York Times suggested that Meta is planning to quietly roll out facial ID in its AI glasses, in order to enhance connection between users of the device. The report, which was based on leaked internal communications from Meta, suggested the company is looking to launch the update amid broader political turmoil in order to get this tool through with limited resistance. But many users are rightfully concerned that such technology could lead to harmful impacts, because people could unwittingly share personal information with glasses wearers. That, according to advocacy groups, could lead to dangerous situations in many contexts, which is why this new coalition is calling for Meta to halt the rollout until more controls can be implemented. Though Meta would prefer to push ahead. The company is looking to advance its AI plans as fast as possible, in order to take on rising competition in the space. As reported by Politico, Meta has already sought to reduce U.S. regulatory rules on AI development through direct consultation with the White House, with a view to ensuring that the U.S. more broadly remains
Apr 13, 2026 · via socialmediatoday.com
Privacy: Civil rights warn against facial recognition in Meta Smart Glasses In an open letter, US civil rights organizations see the right to privacy threatened by Meta's plans for Ray-Ban and Oakley glasses. The plans of the US company Meta to equip its smart glasses with facial recognition represent an unacceptable threat to privacy and freedom and are “a red line that society must not cross.” This is warned by a broad alliance of organizations led by the US civil liberties organization American Civil Liberties Union (ACLU) in an open letter to Meta CEO Mark Zuckerberg. The letter was signed by 75 non-governmental organizations of various stripes. In the letter, they call on Meta to immediately stop its plans to equip its intelligent glasses with facial recognition and to publicly distance itself from them. “Glasses with facial recognition technology would enable any wearer to identify strangers in their vicinity – for example at protests, in doctors' offices or shops – by name. This name could then be linked to digital databases containing a wealth of sensitive information about the target person's profession, habits, health, and relationships,” it says in a press release issued by the ACLU on Monday issued by the ACLU. The US population has not consented to such a massive violation of privacy, according to Kade Crockford, Director of Technology and Justice Programs at the ACLU of Massachusetts. “Stalkers and fraudsters would have an easy time with this technology,” he warns. “Federal officials could use it to harass and intimidate critics. It is dangerous and dystopian, and Meta must distance itself from it.” Minorities particularly at risk The signatories of the open letter see particularly vulnerable population groups, including people of color, women, children, immigrants, religious minorities, and members of the LGBTQ+ community, as being endangered through facial
Apr 13, 2026 · via heise.de
- April 13, 2026 Loading Hospital: Naples Comprehensive Health Size: 703 beds Budget: $1 billion Technology: GI Genius Naples Comprehensive Health began utilizing Medtronic’s GI Genius intelligent endoscopy module soon after it gained FDA approval in 2021. The AI-based image-recognition software runs seamlessly in the background during colonoscopies, examining every image frame by frame in search of abnormal lesions or patterns in the lining of the colon that might be missed by the human eye. “It brings our attention to it by circling it or drawing a little box around it,” says Dr. Kunal Suryawala, a gastroenterologist at NCH. “So then we can look closer in that area to determine if it’s a polyp or area of concern and if it’s something that we need to biopsy or resect. “The purpose of doing a colonoscopy is to decrease the risk for developing colon cancer, and one of the ways we do that is by finding polyps,” he continues. “Not all polyps turn into cancer. But if you have precancerous lesions, we want to identify them and resect them as soon as we see them to decrease the risk of that polyp ever turning into a malignancy down the road.” Use of the software is up to provider discretion, but it’s incorporated into most colonoscopies at NCH. “We’ve done thousands and thousands of cases using GI Genius,” says Suryawala. NCH doesn’t have its own data yet on the impact of the technology. But Medtronic says GI Genius has been shown to increase adenoma (or precancerous polyp) detection rates (ADR) by up to 14.4%, and that each 1% increase in ADR decreases the risk of interval colorectal cancer (cancer diagnosed within five years after a negative colonoscopy) by 3%. “It’s not meant to replace an endoscopist or gastroenterologist,” says Suryawala. “But it
Apr 13, 2026 · via businessobserverfl.com
SAN FRANCISCO — Meta’s reported plan to equip its artificial intelligence (AI) eyeglasses with facial recognition technology poses an unacceptable threat to privacy and liberty and is “a red line society must not cross,” a broad coalition of organizations led by the American Civil Liberties Union, the ACLU of Massachusetts, and the New York Civil Liberties Union (NYCLU) warned today in an open letter to Meta CEO Mark Zuckerberg. The letter was signed by 75 local, state, and national organizations that advocate for domestic violence survivors, worker rights, bodily autonomy, consumer privacy, civil rights, and civil liberties. They called on Meta to immediately halt and publicly disavow plans to equip its Ray-Ban and Oakley glasses with facial recognition. Glasses equipped with facial recognition technology would allow anyone wearing them to identify by name any strangers in their vicinity — including at protests, medical clinics, and businesses. They could then link that name to digital databases containing a wealth of sensitive information on the target's job, habits, health, and relationships. “The American people have not consented to this massive invasion of privacy,” said Kade Crockford, director of technology and justice programs at the ACLU of Massachusetts. “Stalkers and scammers would have a field day with this technology. Federal agents could use it to harass and intimidate their critics. It’s dangerous and dystopian, and Meta must disavow it.” The open letter raises serious concerns about the technology’s impact on vulnerable populations, including people of color, women, children, immigrants, religious minorities, and members of the LGBTQ+ community. Anti-industry activists could use Meta’s glasses to identify and harass corporate executives. Public employees could be identified and harmed because of who they work for, or the political positions taken by their bosses. By eliminating the expectation of privacy or anonymity in public, Meta’s glasses would
Apr 13, 2026 · via aclum.org
ACLU and 75 Organizations Sound Alarm on Meta’s Plan to Add Facial Recognition Technology to Ray-Ban and Oakley Eyeglasses Open letter signed by scores of nonprofits warns that the glasses pose an unacceptable threat to our right to privacy and must be cancelled Meta’s reported plan to equip its artificial intelligence (AI) eyeglasses with facial recognition technology poses an unacceptable threat to privacy and liberty and is “a red line society must not cross,” a broad coalition of organizations led by the American Civil Liberties Union, the ACLU of Massachusetts, and the New York Civil Liberties Union (NYCLU) warned today in a letter to Meta CEO Mark Zuckerberg. The letter was signed by 75 local, state, and national organizations that advocate for domestic violence survivors, worker rights, bodily autonomy, consumer privacy, civil rights, and civil liberties. They called on Meta to immediately halt and publicly disavow plans to equip its Ray-Ban and Oakley glasses with facial recognition. Glasses equipped with facial recognition technology would allow anyone wearing them to identify by name any strangers in their vicinity — including at protests, medical clinics, and businesses. They could then link that name to digital databases containing a wealth of sensitive information on the target’s job, habits, health, and relationships. “The American people have not consented to this massive invasion of privacy,” said Kade Crockford, director of technology and justice programs at the ACLU of Massachusetts. “Stalkers and scammers would have a field day with this technology. Federal agents could use it to harass and intimidate their critics. It’s dangerous and dystopian, and Meta must disavow it.” The open letter raises serious concerns about the technology’s impact on vulnerable populations, including people of color, women, children, immigrants, religious minorities, and members of the LGBTQ+ community. Anti-industry activists could use Meta’s glasses to
Apr 13, 2026 · via aclu.org
A trial of controversial live facial recognition technology that scanned thousands of faces of passengers travelling from Dublin to Holyhead produced no matches, the UK Home Office has said. Thousands of passengers on the ferry route to the Welsh port were scanned in February as part of a UK immigration enforcement pilot scheme. Over three days, passengers were checked against a watch list of 6,535 suspected immigration offenders, with no matches detected. The outcome raises fresh questions about the effectiveness and proportionality of the technology, which Irish and UK authorities are planning to deploy more widely. Despite no passengers matching the watch list, two people were arrested during the operation. The Home Office did not say how the arrests were linked to the facial recognition system. READ MORE A further trial of the powerful surveillance technology was carried out by police forces in England and Wales over six days in November. Across all the trials since then, more than 10,000 faces were scanned against watch lists that grew from 1,942 to 6,535 individuals. The initiative has resulted in just two alerts overall, with none arising in the most recent leg. The operation is a “proof-of-concept pilot” by the Home Office’s immigration enforcement division, which plans to use the technology to locate people within their “Population of Interest”. The Home Office said the six-day pilot in November cost £50,000. A Home Office spokesman said they would not comment on operational matters, but that live facial recognition technology was an essential part of “safeguarding the integrity of the UK’s immigration system”. Olga Cronin, a senior policy officer at the Irish Council for Civil Liberties, questioned the proportionality of the programme. “How can it be necessary and proportionate to subject more than 10,000 individuals with no connection to wrongdoing to indiscriminate biometric facial
Apr 13, 2026 · via irishtimes.com
This article has been disseminated on behalf of MindBio Therapeutics Corp. and may include paid advertising. MindBio Therapeutics (CSE: MBIO; Frankfurt: WF6; OTCQB: MBQIF) a biotechnology company focused on AI-driven voice analytics for intoxication detection, announced the completion of its Intox Collect(TM) technology, enabling rapid scaling of data collection and analysis across a wide range of substances using voice and facial recognition. The platform enhances the company’s ability to train its AI model, which leverages more than 50 million data points to predict drug and alcohol intoxication through voice with high accuracy, while supporting expansion into regulated industries such as mining, aviation and construction, and advancing toward commercialization with upcoming Edge AI kiosk devices expected to begin testing in the second quarter of 2026. To view the full press release, visit https://ibn.fm/8v5XE About MindBio Therapeutics Corp. MindBio is a biotechnology company that is commercialising AI prediction technologies for drug and alcohol intoxication detection via voice analysis. The AI prediction model uses over 50 million data points to predict alcohol intoxication with remarkable accuracy, just by using the human voice. The Company is developing an enterprise platform including Edge-AI kiosks integrating bespoke hardware and software for the detection of drug and alcohol intoxication using the human voice and AI in a range of enterprise environments including the mining industry, aviation, construction and law enforcement. NOTE TO INVESTORS: The latest news and updates relating to MBQIF are available in the company’s newsroom at https://ibn.fm/MBQIF About TechMediaWire TechMediaWire (“TMW”) is a specialized communications platform with a focus on pioneering public and private companies driving the future of technology. It is one of 75+ brands within the Dynamic Brand Portfolio @ IBN that delivers: (1) access to a vast network of wire solutions via InvestorWire to efficiently and effectively reach a myriad of target
Apr 13, 2026 · via markets.financialcontent.com
GenAI fraud makes zero-knowledge proofs non-negotiable By Jarek Sygitowicz, Co-Founder and Chief Strategy Officer of Authologic By now it’s something of a cliché to say the rapid rise of generative AI has fundamentally changed the threat landscape of the internet. Nevertheless, it has, and the point bears repeating. Synthetic identities, deepfakes and AI-generated documents are no longer edge cases but the go-to tools of fraud. The more convincing generative models become, the faster they sideline traditional methods of digital verification like uploading photos of IDs, sharing personal data or performing visual checks. The more data people share, the more material fraudsters have to exploit. Identity verification methods still depend on oversharing. Few flaws are more pressing. Zero-knowledge proofs (ZKPs) offer a way out of this trap. Instead of asking users to reveal sensitive information to prove eligibility, ZKPs allow them to prove a fact about themselves without disclosing the underlying data. When AI can fabricate a face or a document, a photo becomes largely useless as a verification method. More than that, ZKPs limit data exposure. Once viewed as a privacy improvement, data minimisation is now a security imperative. Oversharing as a systemic vulnerability Most online verification flows today unintentionally collect hordes of unnecessary information. Age checks often require a full photo of a government-issued ID, collecting addresses, height and weight, and eye color in the process. Address verification exposes home locations. Identity checks routinely capture names, document numbers, and biometric images, all stored beyond the reach and control of the end user. When fraud required human effort and manual forgery, this model provided adequate protection. GenAI has made this model dangerously outdated almost overnight. A single leaked ID image can be reused and altered at scale. AI systems trained on real documents can generate convincing counterfeits that outpace any
Apr 13, 2026 · via biometricupdate.com
Abstract Existing neural radiance field (NeRF) 3D reconstruction methods have demonstrated significant potential in scene modeling. However, although NeRF shows good performance in 3D reconstruction, it still faces practical challenges when processing remote sensing images captured by Unmanned Aerial Vehicle, such as differences in geometric properties between the height and horizontal dimensions and insufficient efficiency in ray sampling. To address this, this paper proposes a novel UAV remote sensing scene reconstruction framework, HF-NeRF, which incorporates a Height-Enhanced Representation (HER) module. This module performs independent encoding and MLP mapping for the height dimension z, thereby capturing geometric features across different height layers. In parallel, a Frequency-Adaptive Sampling (FAS) approach, which analyzes energy distribution in the frequency domain, adaptively modulates sampling density to improve computational performance and preserve fine-grained details. Experimental results on both a proprietary UAV remote sensing dataset and publicly available datasets demonstrate that the proposed reconstruction approach excels over leading NeRF derivatives in overall performance for remote sensing imagery. Similar content being viewed by others Data availability The raw data supporting the findings of this study include: (1) The Sydney Opera House and High-rise scenes were obtained from the outdoor multi-modal dataset (OMMO), publicly available at: DOI: https://doi.org/10.1109/ICCV51070.2023.00695; and (2) additional self-collected datasets, including The Pavilion, Playground and House scenes, were captured and annotated by the authors, which are available from the first author upon reasonable request via email: 2312391076@st.gxu.edu.cn. References Colomina, I. & Molina, P. Unmanned aerial systems for photogrammetry and remote sensing: A review. ISPRS J. Photogramm. Remote Sens. 92, 79–97 (2014). Nex, F. & Remondino, F. Uav for 3d mapping applications: A review. Appl. Geomat. 6, 1–15 (2014). Eisenbeiß, H. UAV photogrammetry. Ph.D. thesis, ETH Zurich (2009). Shan, J. & Toth, C. K. Topographic laser ranging and scanning: principles and processing (CRC press, 2018). Wallace, L.,
Apr 13, 2026 · via nature.com
The Supreme Court on Monday agreed to examine whether biometric and facial recognition technology could be introduced for voter verification before ballots are cast, with the aim of improving electoral integrity and preventing fraud. It issued notices to the Centre and the Election Commission of India (ECI) on a public interest litigation (PIL) seeking such measures. A bench led by Chief Justice of India Surya Kant and Justice Joymalya Bagchi said the petition, filed by advocate Ashwini Upadhyay, raised important issues requiring consideration. However, it made clear that any such system could not be introduced immediately and would require legal changes. “The prayers cannot be considered for the upcoming elections. But whether such a recourse deserves to be followed for the next parliamentary elections or State polls needs to be examined. Issue notice,” the court said. The bench noted that introducing biometric and facial authentication at polling stations would involve more than administrative action, requiring amendments to existing rules and substantial expenditure. During the hearing, the court initially suggested that Upadhyay first approach the ECI. It observed that the Commission should respond first and noted that implementation would also depend on support from state governments and budgetary approval. The court later agreed to consider the matter after Upadhyay clarified that he was not seeking implementation for ongoing or upcoming Assembly elections, but for future polls. Appearing in person, he said any such reform would need cooperation from state governments. The PIL cites concerns over alleged electoral malpractice, including bribery, undue influence, impersonation, duplicate voting and so-called ghost voters. It seeks directions for the ECI to introduce fingerprint and iris-based verification at polling stations to ensure only genuine voters cast ballots and to uphold the principle of one citizen, one vote. The petition states that current identification methods, based mainly on
Apr 13, 2026 · via outlookindia.com