GetReal Security (@GetRealSecurity) header
GetReal Security (@GetRealSecurity) profile picture

Real security for AI-fueled deception and malicious digital media.

Joined August 2025
“I would comfortably say with less than a minute, you can clone somebody. Instead of needing thousands of images of somebody with different head poses, lighting and camera angles to capture their facial geometry, all that is needed is a single photo to create an avatar deepfake.” Following the federal indictment of three individuals charged with impersonating executives and lawyers to swindle $130 million from enterprises, GetReal co-founder and chief science officer Dr. Hany Farid spoke with the @nytimes about how generative AI has lowered the bar for identity-based attacks. Photos, videos, and audio clips that are normal artifacts of being an executive are now weaponized, often using the same or similar tools available to consumers that can run on consumer-grade hardware. What used to require thousands of images now takes one. That changes what proof is: seeing and hearing a familiar executive is no longer evidence that it’s really them. Closing that gap by verifying identity in real time, across the voice and video channels where these attacks land, is the problem enterprises now have to solve.
1
1
54
Three people, readily available AI tools, and $130 million. A trusted face or voice on a call is no longer enough to defend against executive impersonation. As reported by @ABC News, Federal prosecutors in Manhattan this week unsealed an indictment alleging that three people used AI to impersonate corporate CEOs and attorneys. The playbook showed a clear and repeatable pattern: An email that appeared to come from the CEO asked an employee for help with a confidential financial transaction. A phone or video call followed, with AI mimicking the CEO's voice and, in some cases, their face. The impersonated CEO then directed the employee to attorneys at large international law firms who were supposedly working on the deal. Those lawyers were impersonated as well. The final step was a wire transfer. The victims named in the indictment span industries: hospital systems in Missouri and California, a bank in Rhode Island, a mutual fund company in Massachusetts, a pet food company in Florida, and a U.K. fintech. All of them lost millions.
1
14
“We spent decades securing the network perimeter. Now we have to secure the trust perimeter — every meeting, interview, and interaction where an adversary can impersonate someone worth believing.” GetReal CEO Matthew Moynahan has spent 30 years working in cybersecurity, and he sees the Binance hack — the largest cryptocurrency theft on record — as a starting point for what’s to come. It’s a new era that will be defined by attacks that happen in broad daylight. The pixels and soundwaves of everyday business are now the front door.
1
1
13
“Everyday people shouldn’t have to act like forensic investigators just to browse the internet. That burden belongs with the technology itself.” GetReal co-founder Dr. Hany Farid recently discussed transparency — and trust — in the age of AI with Google DeepMind’s Pushmeet Kohli. When anyone’s face, voice, or writing can be synthesized, the answer can’t be to ask people to squint at pixels. Detecting what’s real must be built into the tools and platforms where media actually shows up — automatically, and in ways anyone can understand in seconds. Farid and Kohli discuss how that happens: watermarking like SynthID, open provenance standards like C2PA, and forensic detection. They aren’t competing approaches, but layers that back each other up. No single signal solves this. The solution is layered, open, and invisible to the people it protects.
1
23
GetReal Security retweeted
Cybersecurity expert Tom Cross, head of threat research at GetReal Security, explains how deepfakes make it possible: AI-enabled recorded or live audio and video that looks and sounds legitimate. Listen to the full Security Management Highlight episode: bit.ly/3UEYdWw
1
1
212
GetReal Security retweeted
“Where we start is not on detection. It’s on the attack.” Dr. Hany Farid, co-founder at GetReal Security and Professor of Computer Science at Dartmouth College, breaks down GetReal’s foundational approach to detecting AI-powered deception.
1
2
42
“AI is erasing the tell-tale signs of a scam.” In the age of generative AI, the clumsy tells that once marked attempted online fraud have given way to convincing deepfakes that can replicate trusted faces and voices. Publicly available media — a social media post, a YouTube video — provide enough information for malicious actors to produce facsimiles of family members, coworkers, and political leaders. @INTERPOL_HQ has joined a chorus of recent warnings from agencies across the globe with a new public awareness campaign that highlights the risk of AI-fueled, identity-based fraud. The campaign provides warning signs and advice for catching these scams before they do harm. Agencies like INTERPOL continue to ring the alarm bell, and in many cases, everyday workers are the ones facing the burden of determining what’s real and who is really on the other side of the call. The recent spate of warnings, advisories, and PSAs are a reminder that while awareness is critical, human perception alone is no longer a reliable safety net. Detection technology that’s built for these emerging threats must be part of the solution.
1
25
One of the things we’re asked the most about at GetReal Security is how to reliably tell if a deepfake attack is occurring. If you’re only relying on human perception — what you see and hear — the answer is: you can’t. Generative AI alongside increasingly sophisticated criminal behaviors, often state sponsored by North Korean operatives, has made it impossible for humans to use perception alone to detect these attacks. October is #CyberSecurityAwarenessMonth, with a focus on how to secure critical infrastructure for the future. Digital identity is the new vector, and for enterprises these attacks create an urgent threat to their ability to do business. The solution is to embed detection technology and security processes along the entire workforce perimeter. Read our guide to five of the most common deepfake tells that aren’t actually reliable. hubs.la/Q04BrfFt0
28
Most people falsely believe they can spot a fake, but the evidence says otherwise. The @AP and @frontlinepbs built a quiz to test that assumption. GetReal Security was one of the two firms they worked with to demonstrate how convincing manipulated media has become. The tells that human perception could once rely on are disappearing. Unnatural blinking, the too-perfect faces, extra fingers, robotic voices — the red flags people were taught to look for are becoming invisible with each new model. Perception alone can't be the defense. Verifying authenticity matters more than how convincing it looks. Take the quiz and see how you do: pbs.org/wgbh/frontline/artic…
12
One of Italy's largest banks was defrauded of roughly €95 million after attackers impersonated the people its executives trusted most, including the use of AI to clone a lawyer's voice. As reported by @Reuters, it began with the chairman of a private bank receiving a WhatsApp message appearing to come from the CEO of its parent company, requesting urgent help with an overseas transaction. A follow-up phone call — purportedly from a senior partner at a prominent law firm — served as confirmation. 🧵
1
27
But Reuters' sources say that voice was actually AI-generated. Backed by forged documents and a fabricated power of attorney, the executive authorized transfers totaling about €95 million (over $107 million USD) to overseas accounts. While the majority of the money was recovered, tens of millions are still gone, routed abroad and converted into cryptocurrency. Prosecutors say this scheme isn’t isolated — they're examining nearly identical scams using the same lawyer's name. Every traditional control relied on to establish trust was a signal attackers are now able to forge. A familiar platform. Official-looking documents. A recognized signature. And the one thing meant to remove all doubt — a live phone call with a voice the executive knew — was the very thing AI was used to counterfeit. The confirmation step became the attack.
1
13
This is the shift financial institutions, and every organization relying on traditional verification checks, are coming to terms with in the GenAI era: authenticating an account or a channel is not the same as verifying the human on the other end. When a familiar voice can be synthesized from public audio, hearing the voice of someone you know is not reason enough on its own to grant trust. Defending against this means treating identity as something you verify continuously and forensically — across voice, video, and every channel where high-value decisions get made — not something you assume. Read the full report from Reuters here (gated content): reuters.com/legal/government…
5
"If we're always looking over our shoulder at every email, every text, every voicemail, every phone call, every video call? It's exhausting." That's our co-founder and Chief Science Officer, Dr. Hany Farid, speaking with Ted Koppel on CBS Sunday Morning. When anyone's face or voice can be faked, the instinct might be to stop trusting and scrutinize every message, go back to in-person-only hiring, and retreat from the tools that make modern work possible. This is why Dr. Farid built GetReal. The answer to AI-enabled deception isn't to trust nothing. It's to build the technology that tells you what's real and who is on the other side of the call. Watch the full conversation with @CBSSunday here: hubs.la/Q04zkYDC0
1
2
42
“Where we start is not on detection. It’s on the attack.” Dr. Hany Farid, co-founder at GetReal Security and Professor of Computer Science at Dartmouth College, breaks down GetReal’s foundational approach to detecting AI-powered deception.
1
2
42
GetReal Security retweeted
Our #forgepointfamily portfolio companies @Surefire_Cyber and @GetRealSecurity recently came together for a timely conversation on deepfakes, the evolving threat landscape, and what organizations need to know to protect themselves. surefirecyber.com/deepfakes-…
1
2
61