When AI Meets AI: The Ultimate Plot Twist
Here’s a story that should make every business leader stop scrolling: In early 2024, an employee at UK engineering firm Arup received what seemed like a routine video call from senior management requesting a financial transfer. The employee followed protocol, verified the faces on screen, and transferred $25 million.
Plot twist: Every single person on that video call was an AI-generated deepfake.
But here’s where it gets interesting.
While criminals are using AI to steal millions, companies are fighting back with their own AI. Meet Daisy, O2’s “AI Granny” who’s been deployed to waste scammers’ time by rambling about her cat Fluffy and her knitting projects. She’s kept fraudsters on the phone for 40 minutes at a time, telling meandering stories while they get increasingly frustrated.
Translation: AI isn’t just the problem – it’s also the solution. But only if you’re smart enough to tell the difference.
The Reality Check: We're All Playing AI Detective Now
Let me share what’s REALLY happening out there.
AI images are getting more sophisticated, but so are we. Half of surveyed individuals can now correctly identify AI-generated art – that’s progress!
The challenge? As AI gets better at creating fake content, it’s also getting better at creating believable misinformation.
The Numbers Tell the Story:
- Financial losses from deepfake fraud hit $200 million in Q1 2025 alone
- Meanwhile, 90% of marketers plan to use AI for content creation in 2025
- But here’s the kicker: 70% of people can’t tell real voices from AI-cloned ones
We’re winning the image game, losing the audio game, and completely missing the misinformation game.
The Three Types of AI Deception Every Business Leader Must Know
1. Malicious AI: When Criminals Get Creative
Remember Builder.ai? My AI/Futurist buddy Graeme Codrington pointed me to this beauty. Microsoft-backed unicorn, $445 million raised, claims their AI assistant “Natasha” builds apps automatically. Reality? Natasha was Indian developers pretending to be AI. The company also allegedly created fake deals to inflate revenue before going bankrupt.
Business lesson: If someone can’t show you their AI working live, they probably don’t have AI.
2. Deepfake Corporate Fraud
The Arup case I mentioned in the opening shows how sophisticated these attacks have become. Criminals are targeting CFOs, HR directors, and anyone with access to company funds using deepfake technology to impersonate trusted colleagues and executives.
Business lesson: Any unexpected financial request via video call gets verified through a different communication channel. Period.
3. AI Misinformation (The Government Edition)
This one’s fresh off the press and absolutely wild. RFK Jr.’s “Make America Healthy Again” report – you know, the one that’s supposed to inform government health policy – cited over 500 studies. Problem? At least seven of those studies don’t exist. The report included “OAIcite” markers (dead giveaway for OpenAI usage) and invented papers with realistic-sounding titles like “Direct-to-consumer advertising and the rise in ADHD medication use among children.”
When epidemiologist Katherine Keyes saw her name on a study she never wrote, she told reporters: “The paper cited is not a real paper that I or my colleagues were involved with.“
Business lesson: This isn’t malicious – it’s lazy. People use AI to create content, then publish it without verification. Remember my 20-60-20 rule? That final 20% isn’t optional – it’s where you catch the mistakes that could tank your credibility.
Five AI-Savvy Strategies That Actually Work
1. Master the Sniff Test
Look for patterns: Too-perfect grammar in casual messages? Celebrity endorsements out of nowhere? Investment opportunities with “zero risk”? Your gut knows something’s off before your brain does.
2. Use Technology as Your Sidekick
- AI or Not (aiornot.com) for suspicious images
- Content at Scale (contentatscale.ai) for questionable text
- Reverse image search for anything that seems too convenient
3. Create Your Verification Protocol
Mine’s simple: Trust my gut, use one detection tool, verify through a different channel. High-stakes content gets the full treatment.
4. Train Your Team on Red Flags
- Urgent requests from “executives” they’ve never spoken to
- Perfect grammar in casual communications
- Any financial request that bypasses normal approval processes
- Video calls where people don’t act like themselves
5. Practice Healthy Skepticism
Ask yourself: “Who benefits if I believe this?” It’s the same critical thinking you use when someone offers you a Rolex from their car trunk.
The Plot Twist: AI vs. AI Is Actually Good News
Here’s what fascinates me about the O2 Granny story: Companies are getting creative about fighting back. O2 intercepted over £250 million in suspected fraudulent transactions last year. Banks are using AI to detect fraud patterns. Social media platforms are deploying AI to catch fake accounts.
The same technology creating the problem is solving it.
But here’s the thing – and this is important – 50% of business leaders admit their employees have no training on recognizing AI-generated content.
You can be ahead of half your competition just by taking this seriously.
Your Action Plan: 5 Things to Do Right Now
-
1. Install Your Detection Toolkit Download AI or Not and Content at Scale. Test them on content you know the answer to.
-
2. Create Your Red Flag Checklist Print it. Stick it by your computer. Reference it before acting on suspicious content.
-
3. Set Up Verification Protocols Any unexpected financial request gets verified through a different channel. No exceptions.
-
4. Train Your Team Send them this article. Ask: "Where are we most vulnerable?" Get everyone thinking.
-
5. Practice Smart Skepticism Start asking: "Who benefits if I believe this?" before clicking, calling, or transferring money.
The Bottom Line: Stay Curious, Stay Smart
AI isn’t going anywhere. Companies using it strategically see 120% increases in organic traffic and 36% higher conversion rates. The question isn’t whether to use AI – it’s how to use it intelligently while avoiding the traps.
Think of it like driving. You don’t avoid cars because some people drive recklessly. You learn the rules, stay alert, and develop good habits.
Same with AI. Get AI-empowered. Stay AI-savvy. Keep building.
Your Homework
Pick one detection tool and test it this week. But more importantly, start asking better questions about everything you see online.
Want to dive deeper into AI strategy that protects AND accelerates your business? Let’s chat about how the UPGRADE Framework can help you become both AI-empowered and AI-savvy.
