)
Artificial intelligence might be the most gripping, divisive and revolutionary technological advancement of the 21st century.
While the innovation is still in its infancy, its potential power is clear for everyone to see. Beyond the basic, day-to-day uses most of us see – writing…*ahem*...editing emails, creating memes of your friendship group and so on, lies a darker use case for AI.
The immense possibility of artificial intelligence has already made its way into the financial sector, both as a means to improve efficiency and, unfortunately, as an effective scamming tool. As the technology becomes more prevalent in everyday life, education will become paramount for the world to protect itself against bad actors harnessing AI’s capabilities.
Artificial intelligence scams
The ever-evolving scope of AI models has led to a growing web of potential scams within the finance – and specifically crypto – industry. According to Chainalysis, artificial intelligence played a significant part in the record $17 billion USD worth of crypto stolen throughout 2025.
Did You Know?
Even more noteworthy was the effectiveness of AI scams when paired with digital assets.
‘Our analysis reveals that, on average, scams with on-chain links to AI vendors extract $3.2 million per operation compared to $719,000 for those without an on-chain link — 4.5 times more revenue per scam.’ - Chainalysis.
)
AI-enabled fraud can take several forms:
DeepFakes and Impersonators
If you’ve spent any time on Reels or TikTok, you’ve likely seen AI used to mimic the voice, and even the faces, of well-known people. Ethical qualms aside, this feature has spilled into the dark world of scams. Some fraudsters have used AI to impersonate famous celebs (like Elon Musk) to endorse fake tokens, or in even more extreme circumstances, to jump on video calls as part of a convoluted phishing attempt.
Impersonation isn’t just limited to the ultra-wealthy. However, AI can be used to mimic CEOs or other executives from local workplaces, making it a more difficult landscape to navigate.
Fake AI Algorithms
AI-powered trading has grown in popularity of late, given the technology’s ability to process large sets of data. However, this can also lead to exploitation. There has been a rise in ‘AI trading bots’ that use ‘unique algorithms’ to deliver ‘guaranteed profits’. In actuality, these programs don’t use AI at all and simply funnel money to the scammers, either immediately or as part of a longer-term Ponzi scheme.
Fake AI Coins
Much like the fake trading bots, ‘developers’ will launch AI cryptocurrencies that supposedly utilise the technology in a novel and impactful way. Often, these projects drum up hype through AI-focussed buzzwords on social media, but don’t actually have any real tech or roadmap backing the token. Once/if the price spikes, the developers will often yank their funds from the project, leaving investors with a practically worthless cryptocurrency.
Who can be targeted by artificial intelligence scams?
Artificial intelligence’s growing presence means that anybody can be targeted by AI scams. The ability to pull off complex fraud means even ‘tech-savvy’ investors may be at risk.
Australians have been named by offshore scam centres as popular targets due to their high median wealth and older average population.
Those with a high profile within the crypto industry, especially if they publicly flaunt their digital asset portfolio, can be targeted.
People working in finance, whether for a crypto exchange or bank, can be targets for impersonation and DeepFake scams.
Artificial intelligence scams: NVIDIA DeepFake Case study
Over the last few years, NVIDIA has become closely associated with AI. As one of the fastest-growing companies of the current decade, its production of high-powered GPUs is synonymous with the role AI is now playing in everyday life.
However, this association has put a target on NVIDIA’s back, as the world saw in late 2025, following a high-profile example of an AI impersonation scam.
NVIDIA’s flagship convention, GTC, took place in October ‘25 in Washington, D.C., with the company’s CEO Jensen Huang set to deliver a keynote speech, covering AI, a big Nokia investment and other key ventures. For those in attendance, the event seemed to go off without a hitch – but online, things took a turn.
The keynote speech was livestreamed across several social media platforms, with thousands of viewers tuning in to hear from the CEO of the world’s biggest company.
As the event got underway, a YouTube channel masquerading as ‘NVIDIA LIVE’ began streaming. In a short period, this livestream had amassed over 100,000 people tuning in – a substantially bigger number than the legitimate speech featured on NVIDIA’s official channel.
To make matters more confusing, the NVIDIA LIVE stream was appearing above Nvidia’s actual stream in Google and YouTube search results.
The fake stream, which had a DeepFaked version of Jensen Huang delivering his keynote, revolved around the AI x blockchain connection, spruiking a supposed ‘crypto distribution where you send one of the supportive cryptocurrencies’ by scanning a QR code.
The QR code, naturally, had nothing to do with NVIDIA at all, and lined the pockets of the scammers. YouTube eventually stepped in to shut down the fraudulent video, but not before it had pulled in over $100k USD in crypto payments.
How to spot AI scams and best practices
The example above is just one of many ways AI has been used to manipulate victims into parting with their hard-earned.
Artificial intelligence is becoming so advanced, so quickly, that even the most experienced DeFi users can be at risk.
That said, by practising cyber hygiene and remaining vigilant, users may be better placed to reduce their exposure to AI scams.
Always verify social media accounts. In the NVIDIA case study, the scammers created a YouTube channel called ‘NVIDIA LIVE’. Clicking into this channel would’ve revealed its age (it was only just created) and that it had basically no footprint. By comparison, the legitimate Nvidia YouTube channel has millions of subscribers and hundreds of videos.
Be cautious of pressure to act quickly. The fake NVIDIA keynote implored viewers to only use the QR code, and not to trust other links online. Generally, it’s best to be wary of anyone (‘friend’ or stranger) who feels forceful in making you spend your money.
Use outbound calls when unsure. If you don’t usually receive calls from a team member and they suddenly request one, it’s worth being cautious. This is especially true if they contact you via an unfamiliar platform – for example, calling on Teams when your organisation primarily uses Zoom. When faced with unexpected call requests, a way to verify legitimacy is to decline the inbound call and contact the person directly using known details.
Consider ‘challenge’ phrases. While AI can do a great job of scraping basic details and impersonating voices/faces, it can be foiled by more in-depth knowledge requests. Consider creating a secret ‘safety’ or ‘challenge’ phrase between family, friends and co-workers when dealing with sensitive information to verify identity.
You can also cross-check companies and websites using sources such as Chainalysis, Moneysmart or Scamwatch.
)
More information
Artificial intelligence can sometimes feel like the wild, wild west – it’s new, powerful, and its capabilities aren’t yet fully discovered. While AI scams can be intimidating, arming yourself with education can go a long way to reducing the risk that you or those around you become victims.
Chainalysis is a prominent blockchain data firm that regularly deals with and prevents, cybersecurity incidents involving AI within the crypto scene. Their website covers a broad range of information on AI scams, how to spot them and case studies as we’ve explored above. For more information to help protect yourself, consider some of the additional reading below.
Additional reading: