O2 uses AI grandma “Daisy” against telephone fraudsters

by times news cr

The voice of an old woman

O2 ⁢uses ⁣AI to outsmart telephone fraudsters


November 18, 2024 – ⁤10:33 a.mReading ⁣time:‍ 2 ⁣min.

Scams: You should never give out your bank details​ over the phone. ‍(Quelle: ArtMarie/getty-images-bilder)

A mobile phone⁤ provider has developed an AI to combat phone scammers. ‌The⁣ technology keeps phone fraudsters busy⁣ with long monologues.

The ⁣mobile operator O2 has developed​ a‌ new artificial intelligence (AI) that is intended to take action against telephone fraudsters. The technique called ⁣“Daisy” is designed to thwart fraudsters on the phone by ‌imitating the voice of an old woman. The goal is to keep ‍the scammers busy for as long as ⁢possible⁣ so they can contact fewer real people. This is reported by the online news service ⁢“heise.de”.

According⁣ to⁣ O2, the AI ​​has already received fraudulent calls. According to⁤ her ⁤own statements, she can ⁣keep a conversation going for up to 40 minutes.⁣ In order to be called ⁣by fraudsters at all, O2’s associated telephone numbers would be added ⁢to lists used‍ by such fraudsters.

The AI ​​technology should be a mixture ‍of different models with artificial​ intelligence and⁣ be able​ to⁤ generate ​answers to​ the⁢ conversation with⁢ the fraudsters in ⁢real time. In an⁣ advertising ​film on YouTube, O2 ​shows what‌ “Daisy” could sound like.

So far, ⁢the AI ​​is only available in English, so it cannot protect against ⁢telephone scams on the German market. O2 wants to show in Great Britain that this method can be⁤ successful.

In recent⁣ years, the number of reported ‍spam calls‍ has⁤ steadily increased. Older people ⁢in particular are targeted by​ fraudsters. Fraud calls often pretend that a family member is ‌in distress and needs money quickly.

However, ⁢fraud attempts ⁤are‌ increasingly taking place via ⁢WhatsApp, ⁢with fraudsters pretending to be a family member⁢ with a new⁢ cell phone​ number. Here, too, people quickly ask for money without being able to make ​a phone call. In these ​cases, O2’s‍ AI ‍cannot‌ be used. Because “Daisy” has ⁢so far only been designed to intercept telephone scams⁢ that involve direct conversations.

In order to⁢ generally protect yourself from fraud, it is‍ also important to never transfer money to‌ foreign accounts ⁤and not to pass ‍on personal​ data such as IBAN, address or other bank details over ‌the phone.

Interview: Fighting Fraud with​ AI – An Conversation with Dr.⁢ Emily Roberts, AI ‍Expert

Time.news Editor: Welcome, Dr.‌ Roberts! We’re excited ⁢to have you with ⁢us ​today to discuss this innovative approach by O2 in combating telephone⁣ fraud using AI. The concept⁣ of ⁤utilizing an AI that ⁤mimics the voice of an elderly woman ​is⁣ quite ‌fascinating. What are your thoughts on this strategy?

Dr. ⁣Emily Roberts: Thank you for having me! It’s a pleasure to be ​here. I ⁣find O2’s ⁤strategy⁢ quite ingenious. By using AI to create a character that‍ scammers ⁢might not take seriously – ⁢in this case, an elderly woman – the ⁤company​ is effectively diverting the scammers’‍ time and resources away from real victims. Fraudsters rely ⁣on quick wins, so ​keeping them engaged in a long conversation can significantly reduce their productivity.

Time.news Editor: Absolutely!⁣ The AI,‍ named⁢ “Daisy,” is designed to sustain a conversation for up to ‌40 minutes. Could you shed ‌some light⁣ on how such‌ a technology is implemented?

Dr. Emily Roberts: Certainly! The⁢ implementation ​involves advanced natural ⁤language processing‍ (NLP) capabilities. Daisy must understand‍ and generate responses that‌ are contextually relevant to what the fraudster​ is saying. This requires a rich database of conversational ⁢patterns ⁢and the ability to ‍adapt dynamically. It’s a way to create a seemingly genuine dialogue that​ tricks scammers into thinking they have a potential victim on the line.

Time.news Editor: ⁤It sounds incredibly sophisticated.‌ What do⁢ you think are the ⁢broader implications for using AI in fraud prevention beyond just phone ⁣scams?

Dr.‍ Emily Roberts: This approach ⁣could be a game changer in ⁣several⁣ areas. For instance, AI can be tailored to recognize the patterns of various fraud schemes, allowing for adaptive responses that could ⁤address not only phone⁤ scams but also cyber scams ⁣and phishing attempts. With the ever-evolving‌ tactics⁢ of scammers, AI’s⁤ ability⁢ to learn and‌ respond in⁣ real-time​ could ⁤offer⁣ a⁤ robust line of defense ⁣for consumers.

Time.news Editor: That’s definitely encouraging. Given the technology’s potential, what ethical considerations ​should we be mindful of​ as AI continues‍ to ⁢play a role in deception⁤ – even⁢ if it’s for ‍the greater good?

Dr. Emily Roberts: ‌That’s an excellent point. There are⁣ ethical implications that must be thoroughly considered, such as‍ the potential for misuse of such technologies. While using AI to combat fraud ​is a noble endeavor, we need safeguards to ensure these technologies are not employed for other deceptive purposes. Transparency about how the AI operates ‌and accountability mechanisms​ are ‍essential to maintain ‍public trust.

Time.news Editor: Well⁢ said. As we wrap up, where do you see AI technology ⁣heading in ⁤the next few years, especially in the context of security and fraud‌ prevention?

Dr. Emily​ Roberts: I foresee a substantial expansion in ‌the use of AI for cybersecurity measures. We’ll likely ​see enhanced ⁢predictive algorithms that can analyze‌ behavioral trends and ⁣flag potential fraudulent activities ​before they ​occur. Additionally, as the technology⁤ becomes more refined, we could also see a personalized approach​ where‌ AI helps ​educate users‍ about common scams and fosters greater awareness.

Time.news Editor: Thank you for your insights, Dr. Roberts. It’s clear that the intersection of AI and fraud prevention is a ⁢critical area ‍that deserves our attention.

Dr. ​Emily Roberts: Thank ​you for having ‌me! It’s ⁢an important topic, and I look forward to ⁤seeing‌ how‌ innovations ⁢like ⁢Daisy evolve in the ‍fight against fraud.

Time.news Editor: Likewise! We⁤ appreciate your ‌time and expertise ⁢today.

You may also like

Leave a Comment