Australian cybersecurity firm Apate is running a global operation that turns the tables on online fraudsters: a fleet of nearly 200,000 AI-powered personas that pose as gullible victims, waste scammers’ time, and extract actionable intelligence for banks, telcos, and crypto analytics firms. The company’s bots field hundreds of thousands of calls and chat conversations daily, and its success is measured in part by an unusual metric — the number of expletives frustrated scammers hurl at the bots.
Wasting scammers’ time at scale
Apate founder Dali Kaafar, a former professor at Macquarie University, launched the company in 2023 after spinning the project out of academic research funded by Australia’s Office of National Intelligence. The idea came from a personal experience in November 2021, when Kaafar strung along a scam caller for 44 minutes during a family picnic — a move that amused his children but annoyed his wife. That call, he said, sparked a realization: if he could waste a scammer’s time for fun, technology could do it at scale.
Also read: FASB proposes new guidance on when stablecoins can be treated as cash equivalents
The company’s bots are designed to sound convincingly human, with distinct personas across genders, ages, and personality types, complete with vocal tics, accents, and natural pauses. They are trained on hundreds of hours of real scam-baiting conversations, allowing them to employ counter-strategies that keep fraudsters engaged. The goal is to maximize the time scammers spend on fake leads, reducing the hours they can devote to defrauding real people.
In the six weeks leading up to the end of 2025, Apate’s bots engaged in 600,000 scam calls for Australian telco TPG alone. Kaafar said that effort wasted more than 500 days of scammers’ time, translating to an estimated $13 million in prevented losses. The company now works with most major Australian banks and institutions in the UK, South Africa, and Southeast Asia.
Also read: Harmony to roll back blockchain after ONE exploit, discarding 109,000 transactions
Extracting intelligence from fraudsters
Beyond wasting time, Apate’s bots are built to gather intelligence. They engage scammers on WhatsApp, Telegram, and other platforms, often acting as honeypots to lure fraudsters into revealing their methods, infrastructure, and financial channels. The bots exploit the greed of scammers, who are often part of organized operations that function like corporate call centers.
In the crypto sector, Apate works with a leading blockchain analytics firm — widely believed to be Chainalysis, though the company did not confirm — to track wallet addresses and fund flows. Kaafar said the bots extract hundreds or thousands of new crypto wallet addresses during conversations, providing early warnings about accounts used to collect scam proceeds. This intelligence helps banks and law enforcement act before funds are moved or laundered.
In July, Apate’s bots uncovered a marketplace where brokers solicited verified bank accounts in India, offering commissions of up to 5% paid in USDT for routing scam proceeds. That discovery highlights the company’s role in exposing fraud infrastructure that might otherwise remain hidden.
An AI arms race in fraud prevention
Scammers are also adopting AI, with Apate’s research suggesting that 20% to 30% of scam text conversations now involve AI-generated content. However, Kaafar argues that defensive AI has a strategic advantage in this arms race. Using game theory, he explained that defenders aim to extract intelligence, while attackers must prompt a specific action — a dynamic that mathematically favors the defense.
“If scammers deploy AI, they’re actually deemed to lose the game,” Kaafar said, citing their researchers’ models. That optimism is tempered by the scale of the problem: global scam losses are estimated at $1.24 trillion annually, giving fraudsters ample resources to invest in automation. Still, Apate’s approach offers a scalable, proactive countermeasure that goes beyond blocking calls or educating consumers.
Conclusion
Apate’s AI-driven scam baiting represents a novel front in the fight against fraud, combining time-wasting tactics with intelligence gathering to disrupt criminal operations. By deploying thousands of convincing personas, the company not only protects potential victims but also provides financial institutions and crypto analytics firms with critical data to trace illicit funds. As both scammers and defenders adopt AI, the battle is likely to intensify — but for now, Apate’s bots are making fraudsters work for every dollar, and occasionally dropping an F-bomb in frustration.
FAQs
Q1: How does Apate’s AI scam-baiting work?
Apate deploys AI-powered personas that engage scammers in realistic phone and chat conversations, wasting their time and extracting intelligence about their methods, wallets, and infrastructure.
Q2: What is Apate’s impact on scam prevention?
In a six-week period, Apate’s bots wasted over 500 days of scammers’ time for one telco, saving an estimated $13 million. The company also provides wallet addresses and other intel to banks and crypto analytics firms.
Q3: Is AI being used by scammers too?
Yes, Apate estimates that 20-30% of scam text conversations already involve AI. However, the company believes defensive AI has a strategic advantage in extracting intelligence and disrupting fraud operations.
This article is for informational purposes only and does not constitute financial, legal, or investment advice. The cryptocurrency and fraud-prevention markets are volatile and uncertain. Readers should conduct their own research and consult qualified professionals before making any decisions.
Be the first to comment