As online scams become more sophisticated, banks, telecommunications companies, and government agencies are looking for better ways to identify fraud before people lose their money. Australian startup Apate.ai is taking an unusual approach to this challenge: instead of waiting for victims to report scams, it sends AI agents directly into conversations with suspected fraudsters.
Founded in 2024, Apate develops AI-powered agents that interact with scammers through phone calls, text messages, emails, websites, and social media platforms. These agents adopt fabricated identities and engage suspected fraudsters to collect information that can help organizations identify and disrupt scam operations.
The company’s approach turns traditional fraud prevention around. Rather than focusing only on blocking suspicious phone numbers or investigating financial losses after they occur, Apate aims to gather intelligence about scammers before they reach more potential victims.
The startup was among the finalists in TechCrunch Disrupt’s Startup Battlefield 200, where it presented its technology and business to investors and the technology industry.
What Is Apate.ai?
Apate.ai is an Australian cybersecurity startup that uses artificial intelligence to investigate scam activity and collect actionable fraud intelligence.
The company was founded by Dali Kaafar, whose research at Macquarie University in Sydney helped establish the technological foundation for the business. Its team developed systems using machine learning and natural language processing to analyze scam communications and improve fraud detection.
Apate deploys AI agents that can communicate with suspected scammers and keep conversations going long enough to gather useful information. Depending on the situation, an agent may answer a scam call, respond to a suspicious message, follow a phishing link in a controlled investigative context, or participate in an online discussion where fraudulent activity is suspected.
The agents can use different conversational styles, languages, accents, and fictional identities to interact with fraudsters.
The information they collect may include phone numbers, bank account details, cryptocurrency wallet addresses, malicious websites, and the identities of brands being impersonated.
Apate then makes relevant intelligence available to customers so they can strengthen their fraud detection systems and respond to emerging threats.
How AI Agents Gather Intelligence From Scammers
Traditional fraud prevention systems often depend on information that becomes available after a suspicious activity has been reported. By that point, scammers may have already contacted numerous people or moved to new phone numbers and online accounts.
Apate attempts to collect intelligence earlier by engaging suspected fraudsters directly.
Its AI agents can operate across several communication channels, including phone calls, emails, text messages, and websites. Text-based agents can also investigate suspicious activity in online forums, investment groups, and social media communities.
During these interactions, the agents try to discover details that can help investigators connect separate scams or identify the infrastructure behind a campaign.
For example, an agent interacting with a suspected investment scammer might attempt to identify a website being promoted, a cryptocurrency wallet being used, or a phone number associated with the operation.
Customers can also assign specific investigative objectives to the agents, such as gathering information about a particular wallet address or bank account.
The resulting information is made available through Apate’s intelligence and analytics portal. Banks, telecom providers, and other organizations can use these findings to improve their monitoring systems, investigate suspicious activity, and support efforts to disrupt fraud campaigns.
The value of the approach depends on the accuracy of the collected information and how effectively customers can act on it.
Why Apate Wants to Waste Scammers’ Time
The idea behind Apate began with a personal experience involving its founder, Dali Kaafar.
According to the company, Kaafar received a call from an unknown number while attending a family picnic. After realizing that the caller was attempting a scam, he decided to continue the conversation rather than immediately hang up.
He reportedly kept the scammer engaged for approximately 44 minutes, entertaining his family while preventing the caller from spending that time contacting other potential victims.
The experience led him to consider whether the same approach could be automated.
If one person could occupy a scammer for several minutes, an automated system might be able to conduct many such conversations simultaneously. More importantly, those interactions could be analyzed to identify patterns and collect information about the fraudsters.
The idea subsequently developed into research at Macquarie University and eventually became the basis for Apate.
The company combines the concept of scam-baiting with AI-driven intelligence gathering. Instead of merely frustrating fraudsters, its agents aim to turn every useful interaction into information that can support broader fraud prevention efforts.
AI Is Making Scams More Difficult to Detect
The growth of generative AI has created new challenges for cybersecurity and fraud prevention teams.
Scammers can use AI tools to produce convincing messages, automate conversations, translate content into different languages, and create more personalized approaches to potential victims. Deepfake technology can also be used to imitate the voices or appearances of trusted people.
These capabilities make it easier to run sophisticated campaigns without the same level of manual effort that such operations once required.
The scale of the problem is reflected in figures cited from the FBI’s Internet Crime Complaint Center. The supplied report states that the FBI received more than 22,000 complaints involving AI-related crimes in 2025, associated with approximately $893 million in losses. More than $632 million of those losses were attributed to investment scams.
These figures highlight why organizations are exploring additional methods of identifying fraud before financial damage occurs.
Apate’s strategy is to use AI against the same broader trend: if automated tools help scammers reach more people, automated investigative agents may help defenders collect intelligence at a comparable scale.
However, AI-driven fraud detection is not a complete solution on its own. Organizations still need effective verification processes, secure systems, human investigation, and timely responses to emerging threats.
How Apate’s Technology Works
Apate uses a combination of advanced AI models and its own fine-tuned models to power its agents and analyze the information they collect.
According to Brad Joffe, the company’s co-founder and chief commercial officer, its systems are trained using conversations collected by the agents alongside research conducted by Kaafar and his colleagues.
The technology serves two connected purposes: engaging suspected scammers and extracting useful intelligence from those interactions.
The process can be understood in four stages:
1. Identify suspicious activity: The system looks for potential scams across communication channels and online communities.
2. Engage suspected fraudsters: AI agents use fabricated identities to participate in conversations and encourage scammers to reveal additional information.
3. Extract and analyze intelligence: The collected information is processed to identify details such as contact numbers, websites, financial accounts, and links between suspected campaigns.
4. Share findings with customers: Organizations access the resulting intelligence through Apate’s platform and use it to inform their anti-fraud measures.
This approach allows the company to focus on the networks and infrastructure behind scams rather than relying exclusively on lists of suspicious numbers or reports from individual victims.
Banks and Telecom Companies Are Already Using Apate
Apate’s customers include major Australian organizations, among them Commonwealth Bank of Australia and TPG Telecom, as well as other telecommunications providers, utilities, and government agencies.
The company has reported results that illustrate how its technology is being applied in practice.
According to Joffe, TPG Telecom has diverted nearly 590,000 calls to Apate’s AI agents. This gives the agents opportunities to engage suspicious callers while collecting information that can support fraud investigations.
Joffe also said Commonwealth Bank of Australia had seen a 76% reduction in publicly reported scam losses.
Additionally, an unnamed Australian bank reportedly used Apate’s outbound agents to disrupt 22 scam campaigns.
These figures were reported by the company and should not be interpreted as independently verified proof that the technology alone produced each outcome. Nevertheless, they provide examples of the types of results Apate says its customers are achieving.
For financial institutions, the potential benefit is better intelligence about fraudulent activity targeting customers. For telecom companies, the technology may help identify patterns across scam calls and associated infrastructure.
Apate Raises $8.15 Million to Expand Its Business
Apate raised $8.15 million in seed funding in August, according to the supplied report. The round included Lobby Capital, OIF Ventures, Investible, Concept Ventures, and Baobab Ventures.
The investment brings the startup’s total funding to just under $10 million, according to Joffe.
Apate plans to use the funding to develop its product further and expand into international markets. Its plans include opening an office in London and building a larger presence in the United States.
International expansion could help the company reach financial institutions, telecommunications providers, and public-sector organizations facing similar fraud challenges.
However, entering new markets may also require the company to adapt to different regulatory requirements, data protection rules, languages, and fraud patterns.
The startup will need to demonstrate that its intelligence remains accurate and useful across different regions as it expands.
What Makes Apate Different From Traditional Fraud Prevention?
Many existing fraud prevention systems focus on blocking suspicious phone numbers, detecting unusual transactions, filtering malicious messages, or investigating incidents after victims report them.
These methods remain important, but scammers can change their contact details, create new websites, and move between platforms.
Apate takes a different approach by engaging suspected fraudsters to collect information about their activities.
Its technology is designed to help organizations understand the broader structure of a scam, including the websites, financial accounts, contact numbers, and impersonated brands associated with it.
That intelligence could help organizations identify related incidents and develop more targeted interventions.
At the same time, the system must distinguish reliable evidence from misleading information supplied by scammers. Investigative activity also needs appropriate safeguards to avoid exposing sensitive data, interfering with legitimate communications, or creating unnecessary risks.
Human oversight and verification therefore remain important parts of an effective fraud prevention strategy.
Conclusion
Apate.ai is using AI agents to change how organizations investigate online scams. By engaging suspected fraudsters directly, the company aims to gather intelligence that can help banks, telecom providers, and government agencies identify and disrupt fraudulent campaigns before more victims are affected.
Its combination of automated conversations, AI-based analysis, and an intelligence portal offers an alternative to approaches that rely primarily on blocking suspicious activity or responding after losses have occurred.
The company’s reported customer results and recent seed funding provide a foundation for its planned international expansion.
As scammers continue to adopt automation and generative AI, the ability to collect timely, actionable intelligence could become increasingly important. Apate’s next challenge will be to demonstrate that its approach can scale across markets while maintaining the accuracy, security, and reliability required for effective fraud prevention.