Apate Sends AI Agents Undercover to Fight Scammers

Most companies deploy AI Agents to ship products faster or streamline workflows. Apate.ai sends its AI Agents undercover to get scammed. The Australian startup has built a fleet of thousands of AI Agents that masquerade as innocent scam victims, according to TechCrunch, which first reported the story on October 14, 2026. The twist is that the company is on the victims' side: every minute a fraudster spends talking to a bot is a minute stolen from a real target.

Apate's AI Agents answer scam calls, reply to shady messages and emails, follow phishing links, and join social media groups to bait fraudsters. They adopt varied personas across ages, genders, accents, and languages, keeping scammers engaged while extracting actionable intelligence. The company says the fleet has so far wasted 17 years of scammers' time across 2.5 million conversations, TechCrunch reports. More importantly, the agents collect the details that are usually only learned after someone has been victimized: bank account numbers, crypto wallet addresses, phone numbers, malicious websites, and the brands being impersonated.

How the undercover fleet works

Apate combines frontier large language models with its own fine-tuned systems, training on conversations its AI Agents have had with scammers as well as the academic research conducted by founder Dali Kaafar at Macquarie University in Sydney, where the company was founded in 2024. The agents talk to tens of thousands of scammers every day across calls, messages, email, and the web.

Customers get more than a wasted-time operation. Apate feeds the gathered intelligence through an insights portal offering analytics on scams targeting a customer's end users, employees, brands, and products. Clients can assign the AI Agents specific missions, such as obtaining a bank account number or a crypto-wallet address. Text agents patrol forums, investment groups, and social platforms for potential scam activity, then enter conversations under fabricated identities to surface fraudsters' details. Co-founder and chief commercial officer Brad Joffe explained to TechCrunch that while most of the industry works after the fact, Apate goes to the scammer directly and gathers information before anyone gets hurt.

The approach flips the standard defense posture on its head. Instead of waiting for a suspicious transaction to trigger an alert, the agents proactively engage suspected fraud operations and return structured evidence that banks, telecom companies, utilities, and government agencies can use to inform anti-fraud measures and initiatives.

Enterprise results so far

The startup's customer roster includes Commonwealth Bank of Australia, TPG Telecom, and other utilities, telcos, and government agencies. According to TechCrunch's reporting, TPG has diverted nearly 590,000 calls to Apate's AI Agents, Commonwealth has seen a 76 percent drop in publicly reported scam losses, and an unnamed Australian bank used the company's outbound agents to disrupt 22 scam campaigns. These figures come from company statements shared with the outlet.

Reported by Digital Market Reports, the company's agents have also engaged scammers in investigating fraudulent operations, collecting information that institutions can act on through an intelligence dashboard. The breadth of channels matters: the same scam operation may run call centers, phishing sites, and social groups simultaneously, and Apate's fleet follows them across all of them.

The deception is deliberately theatrical. Apate takes its name from the Greek goddess of deception, a nod to what its AI Agents do. The company even includes examples of scammers swearing at its bots in pitch materials, treating the frustration as proof of engagement. Every angry exchange is logged as time a fraudster did not spend on a human target.

AI-powered fraud meets AI-powered defense

Apate arrives as AI has dramatically empowered the scammers themselves. The FBI received more than 22,000 complaints of AI-related crimes in 2025, resulting in about $893 million in losses, with over $632 million tied to investment scams where fraudsters use AI to run conversations and deepfake the voices and appearances of celebrities, executives, and trusted figures, as reported by TechCrunch. Agentic defense is emerging as a natural counter: if fraudsters can run conversations at machine scale, defenders can too.

This belongs to a broader shift in how enterprises think about agent safety. Signed identities for AI Agents, like the ServiceNow AI Passport program, and governance platforms that watch for rogue behavior, such as Opsin, address the risk of agents going wrong. Apate addresses the inverse: deploying agents deliberately into adversarial situations where their job is to be fooled on purpose.

Funding, origins, and expansion

Apate raised $8.15 million in seed funding in August from Lobby Capital, OIF Ventures, Investible, Concept Ventures, and Baobab Ventures, according to AI Brief. The 20-person company has raised just under $10 million in total. The startup plans to spend the latest round on product development and international expansion, including opening an office in London and building a larger presence in the United States.

The origin story began at a family picnic, where founder and CEO Dali Kaafar received a scam call and decided to play dumb to entertain his children, keeping the caller on the line for 44 minutes. The experience planted two insights: those were 44 minutes the scammer did not spend on a real victim, and 44 minutes of wasted human time that automation could absorb instead. That curiosity became machine learning research at Macquarie University focused on uncovering the content of scam calls and improving detection, which eventually became Apate.

The company also appeared among the finalists at this year's Startup Battlefield 200 at TechCrunch Disrupt, pitching a panel of VC judges at the San Francisco event this week. Scam-baiting has long been a human hobby; Apate is betting that AI Agents can industrialize it, turning each minute of deception into intelligence that keeps real people safe.