Apate.ai deploys AI agents to engage scammers and collect fraud intelligence for enterprise customers
Australian startup uses thousands of undercover AI agents to waste scammers’ time and extract actionable fraud data for banks, telcos, and government agencies.
1 source · cross-referenced
- Apate.ai, an Australian startup, uses AI agents to engage with scammers undercover, wasting 17 years of scammers’ time across 2.5 million conversations.
- The agents extract fraud intelligence such as bank accounts, crypto wallets, and malicious websites, which Apate provides to customers including banks and telecoms.
- Apate’s customers include Commonwealth Bank of Australia, TPG Telecom, and unnamed utilities and government agencies.
- The startup raised $8.15 million in seed funding in August 2026 and plans to expand internationally, including opening an office in London.
Apate.ai, an Australian startup founded in 2024 based on research from Macquarie University, deploys thousands of AI agents that pose as potential victims to engage with scammers. These agents can answer scam calls, respond to fraudulent messages and emails, follow phishing links, and infiltrate social media groups to bait fraudsters. The agents are designed to mimic a diverse range of people, using varied ages, genders, accents, and languages to keep scammers engaged.
According to the company, its fleet of agents has wasted 17 years of scammers’ time across 2.5 million conversations. The agents aim to extract information typically only revealed after a victim has been scammed, such as bank accounts, crypto wallets, phone numbers, malicious websites, and details about the scams being run. Apate.ai then provides this intelligence to its customers, which include banks, telecom companies, government agencies, and utilities, to inform their anti-fraud measures.
Apate.ai’s approach contrasts with traditional anti-fraud methods, which often act after the fact. As Brad Joffe, co-founder and chief commercial officer, explained, "Most of the industry works after the fact. Telcos block numbers, but scammers just get new ones. Banks watch transactions, but by then the scammer has already reached the customer. We go to the scammer directly and get the information before anyone gets hurt."
The company uses both frontier models and its own fine-tuned models to power its agents and analysis. Apate.ai trains its models on conversations its agents have had with scammers, as well as research conducted at Macquarie University. Customers receive access to an insights portal that provides intelligence and analytics on scams targeting their users, brands, products, and more. Customers can also assign agents specific missions, such as obtaining a bank account number or crypto-wallet address.
Apate.ai’s text-based agents can search forums, investment groups, and social platforms for potential scam activity and then engage with scammers using fabricated identities to surface their details. The startup’s customers include Commonwealth Bank of Australia, TPG Telecom, and other utilities, telcos, and government agencies.
The company reports measurable outcomes for its customers: TPG has diverted nearly 590,000 calls to Apate.ai’s agents, Commonwealth Bank has seen a 76% drop in publicly reported scam losses, and an unnamed Australian bank has disrupted 22 scam campaigns using Apate.ai’s outbound agents.
Apate.ai raised $8.15 million in seed funding in August 2026 from Lobby Capital, OIF Ventures, Investible, Concept Ventures, and Baobab Ventures. The company has raised just under $10 million in total and plans to use the latest funding for product development and international expansion, including opening an office in London and building a larger presence in the U.S.
The startup’s name references Apate, the Greek goddess of deception, reflecting its mission to deceive scammers. Joffe noted that the company even celebrates when scammers swear at its bots, as it views such reactions as a sign of frustration and success.
- Oct 15, 2026 · arXiv cs.AI
Study finds agentic data science systems frequently fail due to methodological flaws despite high predictive performance
Trust79 - Oct 15, 2026 · arXiv cs.AI
Researchers propose ClawLoop, a self-improving agent training framework for harness-native environments
Trust79 - Oct 14, 2026 · Latent Space — swyx
Leash Bio releases Minos, an AI agent sandbox for verifiable scientific reasoning in chemistry
Trust78