Engine

GPT-4o

Meta

Researchers at the University of Illinois Urbana-Champaign (UIUC) have developed AI-driven phone scam agents using OpenAI’s voice API. These advanced tools can execute various phone scams, raising alarms about the potential misuse of artificial intelligence in fraudulent activities. 
According to UIUC assistant professor Daniel Kang, phone scams impact approximately 18 million Americans each year, resulting in losses of around $40 billion. The new AI agents, powered by OpenAI’s GPT-4o model, can mimic conversations and respond to audio prompts, making them more convincing and harder to detect.&nbsp; 
In the research <a href="https://medium.com/@danieldkang/voice-enabled-ai-agents-how-they-can-perform-common-scams-5e3208044b19" target="_blank" data-type="link" data-id="https://medium.com/@danieldkang/voice-enabled-ai-agents-how-they-can-perform-common-scams-5e3208044b19" rel="noreferrer noopener nofollow">report</a>, Kang points out that the average cost of running a successful scam with these agents is a mere $0.75, significantly lowering the barriers for scammers. 
The research team conducted various experiments to assess the effectiveness of these AI scam agents. They focused on common scams like crypto transfers, gift card schemes, and the theft of personal credentials.&nbsp; 
Remarkably, the AI agents had an overall success rate of 36%, with many failures attributed to transcription errors rather than the agents’ capabilities. The simplicity of their design—just 1,051 lines of code—highlights how easily such dual-use technologies can be developed. 
Scammers often impersonate legitimate organizations, such as banks or government agencies, to trick victims into revealing sensitive information. The AI agents can execute complex scams that require multiple steps, such as navigating websites and handling two-factor authentication.&nbsp; 
For instance, a bank transfer scam might involve 26 distinct actions and take up to three minutes to complete. This complexity, coupled with the agents’ ability to maintain coherent conversations, makes them a formidable threat. 
As AI technology continues to advance, users need to remain vigilant. While AI can offer incredible benefits, it also poses risks if misused. Individuals should be cautious about sharing personal information over the phone and remain informed about the tactics employed by scammers. By staying aware and informed, users can better protect themselves against these evolving threats. 
<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow"> 
 Also Read: <a href="https://www.cryptotimes.io/2024/10/29/meta-develops-ai-search-engine-to-compete-with-google/" target="_blank" rel="noreferrer noopener">Meta Develops AI Search Engine to Compete with Google</a> 
</blockquote>

University Researchers Warn of AI-Driven Phone Scams

The Crypto Times

Other articles published on Oct 29, 2024

Dogwifhat’s Trendline Retest: Will WIF Climb to Fresh Peaks?

Bitcoin records a massive rise before US elections : What’s behind it ?

Pepe Musk (PEPEMUSK) Solana Memecoin Will Rally Over 13,000% Ahead of First Exchange Listing, Whi...

Transak Data Breach Exposes 92K Users: Employee Phishing Attack, Ransomware Group Claims Responsi...

EU’s MiCA Regulation: A Potential Crisis for Stablecoins and Crypto Stability

Unlocking the Power of DePINs: A Comprehensive Guide To Top DePIN Platforms [2025]

Join our free newsletter for daily crypto updates!