Biometric Update

Google investing $5M into LatAM DPI projects through Co-Develop

Masha Borak — Fri, 17 Apr 2026 17:14:10 +0000

Google has pledged $5 million in support for Latin American governments adopting digital public infrastructure (DPI), including digital IDs and payment systems. The tech giant cited the example of IdLAC, a digital identity broker that connects digital identity providers with public or private services that need to verify users’ identities. The announcement was made during the World Bank and International Monetary Fund (IMF) Spring Meetings in Washington, D.C. on Wednesday. The funding will go through Co-Develop, a nonprofit fund focused on investing in DPI projects in low- and middle-income countries, and is part of a border cooperation with the Inter-American Development Bank (IDB). First initiated four years ago, IdLAC allows citizens across Latin America and the Caribbean to conduct digital public service transactions in another country using their national digital identity. The service acts as a broker for interoperable digital ID exchange under the Mercosur Digital Citizen initiative, also known as the LAC Digital Citizen or “Regional Digital Citizen.” “This technology is designed to be easily reused, saving nations from building expensive, custom software from scratch,” says Google. “A citizen from Brazil, for instance, can use their credentials in Colombia or Argentina to seamlessly manage entry requirements, verify vaccine coverage, or fulfill medical prescriptions while traveling.” The system was developed by 12 countries in the region through the Inter-American Development Bank (IDB) with support from the World Bank, Co-Develop, the Organization of American States (OAS), and Red GEALC, a network of e-government authorities in Latin America and the Caribbean. IdLAC was launched in October 2025 and piloted for immigration processes for foreign visitors to Chile using Chile's digital identity system ClaveÚnica (Unique Key). “It is estimated that next year most Latin American countries will be able to integrate into this regional digital hub, which could make the region one of the most digitally integrated in the world,” José Inostroza, head of the Chilean Ministry of Finance's Digital Government Secretariat, said at a presentation of the pilot in November. The following month, Uruguay and Brazil announced that cross-border digital identification through IdLAC will be available for foreign trade procedures. Brazilian citizens who have obtained gold-level digital identification through the national platform Gov.br will be able to access services on Uruguay’s Foreign Trade Single Window (Ventanilla Única de Comercio Exterior - VUCE). The software was developed by Uruguayian company Pixys. Among the countries included in development are Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, Ecuador, Guatemala, Paraguay, Peru, Dominican Republic and Uruguay. As part of its partnership with the IDB, Google also plans to offer free online AI training for government officials in the region. In addition, the tech firm has released a report on the impact of AI on Latin American economies. Co-Develop joined the Digital Public Goods Alliance in January to collaborate on the push for safe and inclusive DPI around the world.

Brazil a regional leader

Brazil’s early adoption of IdLAC for foreign trade and Google’s use of its ID system as an example above reflect the progress the country has made on digital government, which the World Bank highlights in a recent blog post. A diagnostic analysis performed at the request of the country’s government by the World Bank’s Identification for Development (ID4D) initiative in 2019 helped shape a broad vision for how digital identity could help the country deliver services online. The result was a leap from 20 million to 170 million registered digital ID users. The challenges faced by the government in implementing the National Digital Government Strategy (ENGD) are analyzed in an article published in the journal Frontiers in Computer Science. Bruno Baranda Cardoso addresses the gaps in digital maturity, risk culture and role of public technology companies that played a role in the development of Brazil’s digital government system.

New version of EU age verification app to follow biometrics bypass, exposure claims

Chris Burt — Fri, 17 Apr 2026 16:33:54 +0000

A new version of the “technically ready” EU age verification app released on Wednesday is expected any time, an official said after commentators online claimed to have discovered privacy and security vulnerabilities in it. Independent developers claimed the vulnerabilities could expose sensitive data stored on user’s devices, including biometrics, and also allow users to bypass the app’s biometric authentication features, a white hat hacker told Politico. The European Commission unveiled the app as ready but not yet available to the public. The developers behind the app, Scytáles and T-Systems, presented it to the age assurance community at the Global Age Assurance Standards Summit. Almost immediately, reports of vulnerabilities from security analysts and began appearing on social media. “DG Connect and the contractor have taken immediate steps,” said EC Digital Spokesperson Thomas Regnier following the revelations. “A new version has just or will soon today be updated. So this is what I mentioned, the code will be constantly updated and improved. It's open source, and I cannot today exclude or prejudge if further updates will be required or not.” Regnier characterized the latest version of the app’s open source code found on GitHub as a demo version. He emphasized that the age verification app’s developers are listening to feedback, and that in the long term it will meet the “highest privacy standards globally.”

SITA urges digital identity, AI coordination as aviation faces ‘significant pressures’

Lu-Hai Liang — Fri, 17 Apr 2026 13:54:40 +0000

SITA’s most recent report mentions the elephant in the room regarding the industry, as the conflict in the Middle East wreaks havoc on global aviation. More than ever, airlines and airports are being urged to close the gap when it comes to data coordination to fully unlock the efficiencies that can be enabled by technology and build foundations that will “outlast the current disruption.” “We are publishing this research at a moment when the industry is under significant pressures,” said David Lavorel, SITA CEO. The executive reveals that across every area measured in the Air Transport IT Insights 2025 report, there arose a common constraint. Where data doesn’t flow freely across systems and partners, investment is unable to fully deliver what it was designed to unlock. “That constraint carries a higher cost today, but also a clear opportunity to emerge stronger,” he said. Airlines and airports are significantly raising their IT spending as they race to build more resilient, data‑driven operations. Airlines invested $36 billion in 2025 — 3.6 percent of revenue — while airports lifted their IT spend to $14.8 billion, or 7.3 percent of revenue. More than 80 percent of both groups now view data‑driven decision making as a strategic priority. The report found operational reliability is closely linked to financial issues, with flight delays alone costing the industry $30 billion annually. Nearly half of airlines (46 percent) are upgrading flight operations systems to unify data across aircraft, crew, passenger and scheduling platforms. However, almost the same proportion (49 percent) point to fragmented data being the biggest barrier to early disruption management. The implementation of the EU’s biometric EES for border control, which involves extensive data-sharing, led to significant delays this week, despite widespread suspension of the biometrics element.

Lack of coordination limits digital identity scale

Digital identity is scaling quickly, with 64 percent of airlines planning to issue their own digital credentials, which is double 2024’s figure, and biometric border controls are expected to reach 83 percent of airports by 2028. But coordination remains the limiting factor as 57 percent of airlines say airport cooperation is vital for digital identity to scale. The International Air Transport Association (IATA) recently concluded that fully contactless international travel is already achievable, enabled by biometrics, if governments begin issuing and accepting digital passports at scale. AI adoption is accelerating, but its impact is constrained by data integration. Among airlines, 63 percent now use AI in operations control to coordinate decisions across aircraft assignment, crew availability and disruption recovery. Generative AI is the top investment priority for the next year. Yet AI is used least in areas that require consistent, multi‑partner data as only 17 percent of airlines use it to monitor turnaround activity in real time. However, airports are closing that gap, with 53 percent now applying it to aircraft turnaround, an increase from 36 percent in 2024. “Aviation is deploying AI with real ambition,” said Lavorel. “But the survey is clear: the primary barrier to maximizing that investment is the lack of data integration across the operation. The technology is there. The data infrastructure to connect it often is not.” Cybersecurity has become a top concern as more systems are interconnected. Cybersecurity is ranked the primary IT focus by 71 percent of airports, and nearly two‑thirds (64 percent) are using AI to detect anomalies earlier. Sustainability is a big focus for passengers worldwide. A global survey from October 2025 by SITA revealed that travelers want faster, smarter and greener journeys. The desire for digital convenience and sustainability was especially pronounced among millennials and Gen Z passengers. However, SITA logs the same pattern in sustainability with airlines advancing in areas they control directly, such as fleet renewal and selective SAF sourcing, while emissions tracking, which requires shared data across operators, remains below 20 percent adoption. Across AI, cybersecurity, digital identity and sustainability, the report finds a consistent pattern: progress is fastest where data flows freely across systems and partners, and slowest where it does not.

US bill would mandate operating system-level age verification

Anthony Kimery — Fri, 17 Apr 2026 09:44:58 +0000

A bipartisan House bill introduced this week, HR 8250, would require operating system providers to verify the age of every user who sets up an account or uses an operating system, shifting age-checking obligations away from individual apps and onto platform owners such as mobile and computer operating system companies. The Parents Decide Act was introduced by Rep. Josh Gottheimer and Rep. Elise Stefanik. “With each passing day, the Internet is becoming more and more treacherous for our kids,” Gottheimer said. “We’re not just talking about social media anymore — we’re talking about artificial intelligence and platforms that are shaping how our kids think, feel, and act, often without any real guardrails.” “Right now, we expect children to self-police their safety online,” Gottheimer continued. “That’s not realistic - and it’s not responsible. Parents should decide what apps their kids can download, what content they can see, and how they interact online - not algorithms or tech companies.” The bill would require users to provide their date of birth to create an account and use an operating system. If the user is under 18, a parent or legal guardian would have to verify the minor’s age. Companies would be required to create a system through which app developers can access the information necessary to verify a user’s age, shifting age-verification infrastructure to the operating-system level. “This approach creates a trusted, consistent standard across platforms,” Gottheimer said. “The phone - the operating system that controls it - will tell the apps and the AI platforms the limits you set for your kid. It gives parents real control, not buried deep in some settings menu, but right in front of them, where it should be.” Gottheimer said the legislation works alongside broader bipartisan efforts to improve online safety, including Sammy’s Law, the Kids Online Safety Act, and the Children and Teens’ Online Privacy Protection Act. The bill also would direct operating system providers to build a system allowing app developers to access information needed to verify a user’s age, subject to Federal Trade Commission (FTC) rules on privacy and data protection. Enforcement would fall to the FTC, which would be required to issue regulations within 180 days of enactment for shared devices, parent verification, and data protection standards to ensure birth-date information is collected securely and not breached. The FTC would be required to brief Congress on its rulemaking process, and within 18 months it would have to submit a report on how providers are complying with the new law and whether Congress should update the requirements. The bill also provides a safe harbor for compliant providers, which might not be held liable under the act if they followed the statute’s requirements and FTC rules. The law would take effect one year after enactment.

NADRA Technologies Limited partners on biometric onboarding, IDV platform

Ayang Macdonald — Fri, 17 Apr 2026 09:29:09 +0000

NADRA Technologies Limited (NTL), the commercial arm of Pakistan’s National Database and Registration Authority (NADRA), has signed a memorandum of understanding with Identity360 Global, a company of EPL Private Lt, to collaborate on the development of a new biometric ID verification system.

A news release announcing the partnership says that the AI-powered tool to be developed is aimed at facilitating identity verification for businesses and other entities in Pakistan and beyond.

Per the release, both parties will "explore opportunities to deploy solutions such as AI-based touchless biometric verification, facial recognition, optical character recognition (OCR – conversion of images into editable digital data), and digital onboarding systems."

Under the agreement, NADRA will combine its well-established national digital ID database, infrastructure, and experience in large-scale identity projects with Identity 360 Global's cutting-edge AI touchless biometrics and liveness detection capabilities.

The AI-based digital ID system, they say, is intended to help prevent fraud, boost regulatory compliance, and facilitate digital know your customer (eKYC) for access to services in different sectors including finance, healthcare, government, and telecommunications.

The expectation is that the new digital ID solution will make a substantial contribution to Pakistan's ongoing financial inclusion and wider digital transformation efforts. It is also expected to strengthen ongoing efforts in digital ID trust.

"Identity360 Global is focused on delivering next-generation AI-based identity solutions. This collaboration with NTL enables us to bring innovative, scalable, and compliant technologies to the market, supporting organizations in building secure digital ecosystems," the CEO of Identity360 Global, Ghazanfar Ali Khan, said of the partnership. He added that the company's work conforms to several international standards and certificates in the area of scalable digital identity.

This partnership agreement with Identity 360 Global adds to a series of moves by NADRA to streamline digital identity management. Last month, the body officially launched the Nishan Pakistan, a unified platform designed to modernize and standardize identity verification services.

Recently, the ID authority also announced an upgrade to the Pak ID mobile app with the addition of easy visa application and facial verification features.

The app has been credited with revolutionizing how Pakistanis access services in the public and private sectors.

AI voice fraud draws new congressional scrutiny

Anthony Kimery — Fri, 17 Apr 2026 09:07:22 +0000

U.S. Sen. Maggie Hassan is escalating congressional scrutiny of the fast-growing AI voice-cloning industry, pressing four major companies to explain what they are doing to stop scammers from turning synthetic speech tools into engines of fraud. In letters dated April 16 to ElevenLabs, LOVO, Speechify, and VEED, the New Hampshire Democrat and ranking member of the Congressional Joint Economic Committee demanded detailed answers about what they are doing to prevent bad actors from using their services. Hassan wants to know whether the companies monitor for scam-related uses, verify that a person has consented before their voice is cloned, detect attempts to imitate public figures and minors, watermark AI-generated audio, preserve provenance information, and report bad actors to law enforcement. The letters amount to more than another general warning about the harms of AI. They reflect a more specific congressional concern that voice models have become highly usable, widely accessible, and increasingly difficult for ordinary people to detect. “In recent years, global criminal networks have used deepfake voice programs, along with other new AI tools, to target more people with increasingly personalized and believable digital scams, fueling a booming scam industry that surpasses the global drug trade as an illicit industry,” Hassan told the companies. “Protecting Americans from these financial losses will require collaboration between the public and private sectors, and AI companies [including yours] are on the frontlines of this effort,” Hassan added. Hassan repeatedly frames the problem in operational terms. She is not only asking whether companies prohibit fraud in their terms of service, but whether they enforce those policies, how often they update scam phrase lists, how many violators they have caught, when they ban users, whether those users can return under new accounts, and whether law enforcement receives information that the public does not. That focus matters because the threat is no longer hypothetical. Hassan pointed out that the Federal Bureau of Investigation’s (FBI) 2025 Internet Crime Report, released this month, shows victims lost $893 million to AI-related scams in 2025, a figure that underscores how quickly synthetic media is being absorbed into familiar fraud schemes. Cryptocurrency and AI-related scams were among the costliest, the FBI said. The FBI also said the Internet Crime Complaint Center received 1,008,597 total complaints, an increase from 859,532 in 2024. Phishing/spoofing, extortion, and investment schemes were the most frequently reported complaints. Americans over 60 reported approximately $7.7 billion in losses, up 37 percent from 2024. Industry and consumer advocates have been warning about the same trend. Consumer Reports said in its March 2025 assessment of AI voice-cloning products from Descript, ElevenLabs, Lovo, PlayHT, Resemble AI, and Speechify that it “found a majority of the products assessed did not have meaningful safeguards to stop fraud or misuse of their product.” Consumer Reports said the platforms should automatically flag and prohibit audio containing phrases commonly used in scams and other fraud, a recommendation that closely tracks the questions Hassan posed in her letters to some of the same companies. Those letters lay out why lawmakers are alarmed. Hassan cited research finding that people are poorly equipped to identify AI-generated voice clones and notes that these systems can create convincing synthetic voices from only a brief audio sample. Hassan highlighted how easy it has become to pick from prebuilt voice libraries or generate synthetic voices in many languages. ElevenLabs, for example, is described as offering thousands of voices in dozens of languages; LOVO more than 500 voices in 100 languages; Speechify more than 1,000 voices in over 60 languages; and VEED more than 35 voices capable of speaking dozens of languages. Hassan said romance scams, impersonation scams, and so-called grandparent scams have manipulated victims into believing a loved one is in danger. She noted a 2025 case involving New Hampshire families who were allegedly tricked by an AI-generated imitation of a relative’s voice, as well as 2024 reports from Merrimack County, New Hampshire, where residents received scam calls from voices made to sound like family members or law enforcement. Voice cloning has also been used against businesses to bypass voice-based authentication or impersonating executives to authorize transfers of large sums of money. Another striking element of Hassan’s inquiry is how directly it targets platform design choices. Several of her questions ask whether the companies require real-time audio for verification, whether they demand authentic non-public recordings before allowing a clone, and what mechanisms they use to determine whether submitted audio is genuine. She also wants to know whether the companies detect when users try to create “no-go” voices, such as politicians and celebrities, and whether they can tell when a user succeeds in bypassing those safeguards anyway. Her letters also probe whether the companies permit the cloning of minors’ voices or the creation of synthetic child-like voices, and if so, what protections they have in place against exploitative misuse. Hassan’s line of questioning dovetails with one of the central critiques of the sector: many voice-cloning products historically relied more on user promises than on meaningful technical guardrails. Consumer Reports said most leading products it examined lacked strong technical mechanisms to stop nonconsensual voice cloning and recommended both identity-focused controls and automatic scam phrase detection. Hassan is asking the companies whether they have gone beyond self-attestation and basic policy language to adopt the kind of systems critics say are necessary. Hassan’s oversight push comes as Congress considers a more formal legislative answer. Senate bill S.3982, the AI Fraud Accountability Act of 2026, would establish a federal framework aimed squarely at digital impersonation fraud. Introduced last month by Republican Sen. Tim Sheehy and Democrat Lisa Blunt Rochester and referred to the Senate Committee on Commerce, Science and Transportation, the bill would amend the Communications Act of 1934 as amended to create a criminal prohibition on using a “digital impersonation” in interstate or foreign communications with intent to defraud someone of money, documents, or anything of value. A companion bill was introduced in the House by Republican Vern Buchanan, vice chairman of the House Committee on Ways and Means and chairman of the House Democracy Partnership, and Democratic Rep. Darren Soto. Both bills define digital impersonation broadly to cover convincingly fabricated or altered audio or visual depictions of either an identifiable real person or even an imaginary person presented as genuine. The bill would authorize penalties of up to three years in prison, include forfeiture provisions and establish extraterritorial federal jurisdiction, a notable provision given that many scam operations originate abroad. “We are seeing a disturbing rise in AI-generated voice clones and deepfake videos that convincingly impersonate loved ones, business executives, government officials, and trusted institutions to steal money,” Buchanan said. “Congress must act to stay ahead of these threats by modernizing federal law to keep up with emerging technology. The AI Fraud Accountability Act makes clear that if you use AI to defraud Americans, you will be prosecuted,” Buchanan added. The AI Fraud Accountability Act would create a civil and regulatory enforcement route through the Federal Trade Commission (FTC). A violation would be treated as an unfair or deceptive act or practice enforceable by the FTC. The bill is structured not only to punish fraudsters after the fact, but also to make digital impersonation fraud a matter of consumer protection enforcement. The bill also contains a standards and governance component. It would require the Secretary of Commerce, acting through the National Institute of Standards and Technology (NIST), to convene a working group within 30 days of enactment to develop best practices for recognition, detection, prevention, and tracing of digital impersonations used in fraud. The working group would include representatives from the Department of Justice, FTC, federal, state, and local law enforcement, private sector industries such as financial services, telecommunications, health care, retail, and digital platforms, as well as scientists and engineers with expertise in digital forensics and AI. NIST would then be required to publish best practices and update them annually. That structure is revealing. Hassan’s letters seek data from the companies about what safeguards exist, how effective they are, and where the gaps remain. The bill, by contrast, tries to build the enforcement and technical architecture that would follow from such findings. In that sense, Hassan’s letters and the AI Fraud Accountability Act are complementary. Hassan is gathering the kind of information Congress would need to judge whether voluntary industry practices are working. The AI Fraud Accountability Act, if law, would supply the beginnings of a statutory answer if lawmakers conclude those practices are not working.

Nearly 40% of Gen Z report fraud losses as scams shift online: TransUnion

Masha Borak — Fri, 17 Apr 2026 08:50:45 +0000

Gen Z is increasingly being targeted by online scammers: Nearly 40 percent of Gen Z consumers reported losing money to digital fraud in the past year. The reason behind this is that members of this generation are more likely to use gambling and betting platforms, social platforms such as forums and dating apps and video games – all of which are experiencing a rise in fraud attempts, according to a new report from TransUnion. The solution to the spread of fraud is increasing identity defenses, the consumer credit agency notes in its H1 2026 Top Fraud Trends Update. Although digital fraud rates are declining overall, more sophisticated fraud schemes, driven by the rise of generative AI and synthetic identities, are causing greater losses for consumers, according to Naureen Ali, U.S. head of fraud at TransUnion. “Addressing this requires a new generation of identity-centric defenses that combine advanced analytics, adaptive authentication and multilayered fraud detection,” says Ali. “Organizations must match fraudsters’ technological innovation to stay ahead of rapidly changing schemes.” The firm found that about one in six U.S. consumers lost money to scams conducted via email, phone calls, texts, or online channels in 2025, with median losses reaching US$2,307. Credit card or fraudulent charges were the leading cause of digital fraud losses, accounting for a third of cases. Identity theft followed closely at 29 percent, while account takeover (ATO) affected one in four (27 percent) victims. TransUnion also published global fraud data, surveying a total of 24 countries. The company found that account creation has become a growing target for fraudsters across the world. Globally in 2025, more than eight percent of account creation attempts were flagged as suspected digital fraud, an 18 percent jump from the year prior. “Instead of bypassing controls during account use, they increasingly exploit vulnerabilities at account creation, concealing identity manipulation until losses mount,” says Ali. The solution is to detect sophisticated identity risks at onboarding, she adds. Over a quarter of global consumers said they lost money to digital fraud, reporting a median loss of $1,671. Money mules and third-party seller scams on legitimate ecommerce sites were the leading causes of loss (24 percent), followed by voice phishing or vishing (23 percent). The report also offers data on other types of fraud, including phishing, smishing, unemployment fraud and social engineering.

Vietnam mandates face biometrics for mobile device registration

Ayang Macdonald — Fri, 17 Apr 2026 07:58:19 +0000

A facial recognition process is now required for new mobile device registrations in Vietnam. The policy took effect April 15 under a circular issued March 31, following a draft released in January, and is intended to combat identity fraud and other illegal activities related to SIM card and mobile device ownership. The registration process entails submitting one's name and date of birth, as well as their national ID number and face biometrics. Authorities say the measure, which is in line with a circular of the Ministry of Science and Technology, will be particularly useful to curb identity theft in situations where individuals lose their devices or where ownership changes without a deactivation of the previous registration. Per the new regulation, telecoms companies have up to two hours to identify activities involving a device change and block outbound services until the owner of the new gadget completes a facial recognition-based registration process. Once there is a block, the affected individual has up to 30 days to undertake the biometric verification to prevent a suspension of both inbound and outbound services. If the situation is not sorted out five days after the full service suspension, the telco is required to put an end to the subscription. The new policy requires not only new SIM cards to be registered using facial recognition, but also identities linked to newly registered devices to be verified against the national population register and the resident database. With the new measure, authorities say it is possible to conduct SIM card registration using the VNeID national digital ID application, and users can also verify the number of SIM cards linked to their ID using the platform. Meanwhile, they can complete the process either in physical offices or through other mobile applications made available by telcos. The new measure is part of the Vietnamese government's efforts towards combating identity theft, something which is also common with social media use. In February, measures aiming to mandate digital ID verification for social media use were announced.

UK social engineering scams jump 62% as fraud tactics shift: BioCatch

Masha Borak — Fri, 17 Apr 2026 06:41:14 +0000

While the United States is battling with credit card fraud and identity theft, UK consumers are being targeted by increased attempts at social engineering attacks, behavioral biometrics company BioCatch warns. According to data from nine UK financial institutions serving more than 100 million accounts, in 2025, social engineering scams increased 62 percent year-on-year. The result is unsurprising, according to BioCatch. Banks have been boosting their technological protections against fraud, leaving scammers with options such as social engineering attacks, which rely on psychological manipulation and deception rather than technical prowess. Remote-access and malware-related fraud attempts both declined, while purchase scams (63 percent), investment scams (34 percent), and romance scams (47 percent) saw upticks, according to the 2026 Digital Banking Fraud Trends in the UK report. Phishing attacks in the UK skyrocketed by 140 percent in 2025, while fraud tied to stolen devices wasn't far behind, climbing by 112 percent. In London alone, more than 70,000 phones were reported stolen in 2025, according to police data. “Stolen devices undermine strong customer authentication,” says Jonathan Frost, BioCatch global advisory director. “To address this, financial institutions should continuously assess behavioral intent, because once a bad actor has control of a trusted device, they can often commit fraud with relatively little friction.” The company concludes that winning the battle against fraud will rely on sharing intelligence and using technologies and tools that identify behaviors, patterns, and signals during transactions and interactions.

AI agent delegation via MCP has gaps a Murderbot could walk through

Chris Burt — Thu, 16 Apr 2026 20:25:33 +0000

The introduction of Model Context Protocol (MCP) open standard developed by Anthropic has advanced the data-sharing capabilities of AI agents and the systems they interact with, but the question of how to secure these interactions from rogue agents and a host of other threats remains open. Gluu Founder and CEO Michael Schwartz presented his vision for secure AI agent authorization in a talk titled “Golem to Murderbot: Challenges with Agentic Security Delegation via MCP” at the MCP Dev Summit 2026 in New York City. The Hebrew story of the Golem from late antiquity raises the question of how an automated actor can be relied on to carry out the intent of the person who automates it. In the story, the Golem becomes unruly as it carries out its task. The “truth” – the equivalent for an AI agent of its mission – becomes more unstable with each change in its network context. Fortunately, a “kill switch” is built into the Golem. In Murderbot, a series of books by Martha Wells adapted into an Apple TV series, the “Corporation” which controls “SecuBots” like Murderbot uses a “Governor Module,” a software module which monitors and “punishes” them for policy violations. The title character has gone rogue and hacked its Governor Module, but must fool a second oversight mechanism, the “Hub System,” that everything is in order by feeding data back to it. Murderbot’s presence is necessary as a security agent to reduce the risk in the scenario the story depicts to the point where it is insurable.

Automation and risk reduction

Schwartz argues that while some people tend to see zero trust in a typical agentic AI flow as a matter of enforcing security at an MCP Gateway, because it is a chokepoint, “we should be good. “But this would imply that all the traffic is trusted beyond the gateway, which is sort of the definition of what zero trust seeks to avoid in the first place,” Schwartz says. Instead, “each service needs a Governor Module,” in the form of a policy engine embedded with each service. Each would then produce decision logs, scaling security data and requiring “more operational leverage” for humans to make use of it to take security actions. Schwartz then explained that human authentication is pretty much solved with mechanisms like passkeys and digital wallets, and even software authentication is for the most part functionally solved. Authorization is another matter. From an enterprise perspective, the question is: “under what conditions is access allowed?” The answer may depend on things that have nothing to do with the properties of an AI agent requesting data on behalf of a human. Schwartz gives the example of data governed by agreements between different organizations. Authorization therefore needs to move beyond role-based access control to policies that include context and complexity. This leads to his case for using Cedar as “a policy syntax that is analyzable.” Schwartz concluded his talk by presenting the concept of GovOps, an operating model for enterprise governance through risk management, accountability and transparency. “Identity is the key for accountability,” Schwartz says, “not authorization.” The presentation is posted to Schwartz Identerati Office Hours channel.