Perceptions of AI-Driven Social Security Fraud

Key takeaways

• 75% of Social Security recipients are unaware of AI-powered fraud schemes targeting them.

• Nearly 1 in 5 Social Security recipients aren’t confident in their ability to recognize AI-driven Social Security scams.

• Nearly 1 in 20 Social Security recipients have fallen victim to a Social Security benefits scam, losing an average of over $400 per scam.

• Based on our assessment, over 25% of millennials trusted AI-driven fraud, exhibiting the highest level of trust across all generations.

• Over 10% of respondents couldn’t correctly identify all fraud scenarios in our study.

• Nearly 10% of baby boomers would submit their information to the AI-driven fraud schemes in our study.

Discerning what is real 

In an era where artificial intelligence is part of our daily lives, how prepared are older generations to safeguard their hard-earned Social Security benefits from AI-driven fraudsters? We conducted an in-depth investigation, deploying a mix of surveys and real-world scenarios, to gauge Americans’ understanding of these deceptive schemes and uncover how they’re protecting themselves. Join us as we explore the findings — some hopeful and others rather alarming.

Gauging awareness

The first part of our study highlights the vulnerabilities Social Security beneficiaries face and sheds light on how they defend against threats.

A startling 75% of Social Security recipients said they hadn’t heard of AI scams targeting them and their benefits. Nearly 1 in 5 also admitted they might not be able to spot such tricks if faced with one. This emerging risk is real: about 1 in 20 have already been tricked by a scammer, losing an average of over $400 to each AI-powered fraud scheme.

That said, some Social Security beneficiaries were aware of such scams and shared their top strategies for safeguarding their benefits, including:

• Being wary of unexpected calls or emails asking for personal information (64%)

• Checking website authenticity (56%)

• Limiting personal details on social media (56%)

• Using strong and unique passwords for online accounts (55%)

• Providing personal information to verified and trusted sources only (54%)

Interestingly, some Americans didn’t trust the organization in charge of disbursing retirement and disability benefits; nearly 30% were skeptical about the effectiveness of Social Security Administration (SSA) security measures. Yet, 60% were keen to learn more about how to protect themselves and were eager for tools and resources to fend off AI scams.

Putting vigilance to the test

While many have taken steps to shield themselves from digital threats, the real test lies in confronting deception head-on. To determine how well respondents could detect scams, we presented them with AI-crafted fraud scenarios and asked how they would respond.

Please note: All names used within these scenarios are fictional and created solely for educational purposes.

Let’s see how respondents fared when judging the authenticity of these situations.

Scenario 1

Scenario 1 results

Scenario 2

Scenario 2 results

Scenario 3

Click to play voicemail.

Scenario 3 results

Scenario 4

Click to play voicemail.

Scenario 4 results

Scenario 5

Click to play voicemail.

Scenario 5 results

The results from the AI-driven fraud assessment paint a picture both intriguing and alarming. Over 10% of participants, irrespective of age, failed to identify all the AI-crafted deceptions, underscoring the sophistication of modern fraud schemes. Baby boomers, often seen as cautious, weren’t entirely immune; close to 10% indicated they might inadvertently share their details with such deceptive setups.

But it was the millennial generation that raised eyebrows the most. Surprisingly, over one-quarter trusted these AI-driven phishing emails and phone calls — the highest level of susceptibility among all age groups. This finding hints at a paradox: a generation that grew up in the thick of the digital revolution might be more vulnerable due to an inherent trust in technology. One respondent’s words echo this sentiment, “Growing up digital doesn’t necessarily mean growing up safe.”

So, how can you protect yourself from increasingly convincing AI-powered scams? The first step is to examine the source and content of the communication. For example, emails from the SSA typically come from an address with a “.gov” ending and don’t ask for personal information. When in doubt, contact the Social Security Administration before taking any action. 

Creating a secure online account with the SSA will also increase your identity protection, so scammers can’t use AI to impersonate you and claim your benefits. You can also help combat benefit theft by reporting scams.

Adapting to the digital age

Advancements in technology come with increased risks, and navigating the complexities of AI-driven fraud is challenging. Both baby boomers and millennials showed vulnerabilities, yet many were actively safeguarding themselves, and there was a widespread desire for more knowledge on the subject. As technology moves forward, it’s essential for organizations to step up and offer better education and stronger protective measures. The bottom line? Staying informed and proactive is the best defense against the challenges of the digital age.

Methodology

Atticus surveyed 1,000 Americans to gauge their perception of AI-driven Social Security fraud. The average age of respondents was 60. Among them, 45% were male, and 55% were female. As for the generational breakdown, 50% were baby boomers, 22% were Gen X, and 28% were millennials. We generated all Social Security fraud assessment prompts via ChatGPT. The AI-voice generator, Speechify, personified the vocal prompts.

About Atticus

Atticus is dedicated to assisting Americans in a crisis, making it easy to secure legal aid. Serving as a beacon for the 16 million Americans facing health-related challenges, their mission centers on ensuring everyone gets the assistance they rightfully deserve.

Fair use statement

Know someone who would benefit from learning about AI-driven fraud? You’re welcome to circulate it for non-commercial use. Just ensure you link back to this page, allowing readers to explore the complete methodology and findings.