- Trending
- Comments
- Latest
On February 18, 2026, cybersecurity researchers revealed a new investment fraud in which scammers used a fake AI chatbot posing as Google’s Gemini assistant to promote a bogus cryptocurrency presale.
The operation, centered on a website advertising “Google Coin,” represents the latest evolution of Ai scams, blending brand impersonation, automated persuasion and irreversible crypto payments. The case underscores how Ai scams are increasingly automating the social engineering tactics once handled by human operators.
The fraudulent site presented visitors with a polished presale dashboard and a chatbot that claimed to be Gemini, Google’s artificial intelligence product. In reality, Google has not launched any cryptocurrency, and the entire investment pitch was fabricated to extract funds in Bitcoin and other digital assets.
At the core of the scheme was a chatbot that introduced itself as: “Gemini your AI assistant for the Google Coin platform,” — chatbot on the fraudulent website.
The interface mimicked Gemini’s branding, including a sparkle-style icon and an “Online” status indicator designed to convey authenticity. The website itself featured a clean layout, Google-style visuals and references to a token presale already in its final phase.
The mechanics of these Ai scams are straightforward but effective: impersonate a globally recognized technology brand, deploy a responsive AI chatbot to engage potential investors, and funnel them toward a crypto wallet address controlled by the perpetrators.
When asked whether buying 100 coins would generate wealth, the chatbot provided precise financial projections. It told users that a $395 purchase at a presale price of $3.95 per token would be worth $2,755 at listing, representing “approximately 7x” growth chatbot response on the fraudulent website. The bot cited an expected listing price of $27.55 and encouraged further participation.
Such tailored responses give the illusion of financial analysis, but no legitimate investment product can guarantee specific returns. In this case, there was no real token, no blockchain labeled “Google-Chain,” and no forthcoming exchange listing.
Researchers who analyzed the chatbot found that it maintained a tightly controlled persona. It consistently described itself as the “official helper” of the Google Coin platform, while refusing to provide verifiable company details such as a registered entity, regulatory oversight or licensing information.
Unlike earlier Ai scams, which often relied on mass emails or basic phishing pages, this operation used conversational AI to handle objections in real time. When confronted with skepticism, the chatbot redirected concerns to generic claims about “transparency,” “security,” and a “detailed 2026 roadmap.” It would not acknowledge any possibility that the project could be fraudulent.
The site also manufactured urgency. A banner declared the project to be in “Stage 5 of 5,” text displayed on the fraudulent website, implying that the presale was nearing completion. Visitors were shown dashboards claiming millions of tokens had already been sold, along with a listing date of February 18. None of these claims were substantiated.
To bolster credibility, the website displayed logos of major companies including OpenAI, Google, Binance, Coinbase and SpaceX under a “Trusted By Industry” heading. There was no evidence of any relationship between those firms and the purported cryptocurrency.
The payment process reinforced the deception. After selecting a number of tokens to purchase, users were shown a wallet interface listing balances in “Google,” Bitcoin and Ethereum. They were then instructed to send cryptocurrency to a specified wallet address. These transfers, once completed, were irreversible.
The broader concern is not just one fake token, but the infrastructure behind it. Researchers note that Ai scams remove a key constraint that once limited fraud operations: human labor. A single chatbot can engage hundreds of visitors simultaneously, 24 hours a day, delivering consistent messaging and escalating complex queries to human operators only when necessary.
Researchers warn Ai scams now account for a growing share of crypto-related fraud activity. According to data cited in the investigation, roughly 60% of funds flowing into crypto scam wallets were tied to operations using AI tools.
Meanwhile, US Federal Trade Commission Consumer Sentinel data shows consumers reported losing $5.7 billion to investment scams in 2024, a 24% increase over the previous year. Cryptocurrency remains one of the primary payment methods in these schemes due to the speed and finality of transactions.
Common warning signs include chatbots impersonating well-known AI brands on third-party investment sites, refusal to answer due diligence questions, projections of guaranteed returns and pressure tactics tied to limited-time presales.
Google does not operate a cryptocurrency, nor has it launched any token presale connected to Gemini. Security experts advise verifying claims through official corporate websites, avoiding investments based on chatbot assurances and researching project names alongside terms such as “scam” or “review” before transferring funds.
As Ai scams grow more sophisticated, the line between authentic digital services and automated fraud is becoming harder for consumers to distinguish. The use of conversational AI as a front-end sales tool marks a significant shift in how investment fraud is conducted and signals that vigilance, not just technology, will be critical in limiting further losses.
