Here’s the thing hardly anyone wants to name out loud: review floods aren’t just “spam.” They’re a denial-of-service attack on reputation, aimed not at your servers, but at the fragile trust graph that decides whether a stranger taps “Call,” “Book,” or “Buy.” One bad afternoon and your Google Business Profile, Yelp page, App Store listing, or Shopify plug-in becomes a surface to overwhelm. Not to hack. To drown.

Call it a reputational DoS.

If you’re a startup founder or an SMB exec, the mechanics look disturbingly simple in 2025. AI makes it cheap to mass-produce convincing, situationally on-brand, grammatically clean reviews; throw-away accounts (or a rental network) provide distribution; a triggering event, or even no event at all, makes the attack legible. Sometimes it’s extortion via gift cards (“pay us or we 1-star you into the ground”), sometimes competitor sabotage, sometimes an ideological pile-on with nothing to do with your product. Meanwhile, platforms are racing to algorithmically skim the scum off the top. It’s a cat-and-mouse game where your cash flow is the cheese. (Leader, 2025; Schwartz, 2024).

What a “review flood” feels like on the ground

The sequence is painfully consistent. A sudden spike, hundreds of 1-star drive-bys or an unnatural geyser of perfect 5-stars. The text reads plausible, even personable. The accounts look human enough. Sales wobble, inbound calls drop, partners start forwarding screenshots, and your team loses a week arm-wrestling platform forms instead of shipping. “Flood” is not metaphorical; it’s volumetric pressure applied to the most public, least resilient system your business depends on: credibility signals in a feed.

The Hidden Cost of Fake AI Tools
How fake ChatGPT apps and extensions exploit speed, trust, and convenience to hijack business accounts and bleed ad spend.
4FshSagger Khraishi

Platforms openly acknowledge the scale. Google says it “blocked or removed more than 240 million policy-violating reviews” in 2024 and “removed or blocked more than 12 million fake Business Profiles,” rolling out alerts that warn customers when suspicious five-star spikes suggest pay-to-play (Leader, 2025). The tone is telling. “An honest five-star review is a great way for customers to show appreciation,” the company writes, before noting that buying five-stars is “strictly prohibited.” (Leader, 2025).

Even with new ML, Google reported taking down 45% more fake reviews in 2023 than 2022, over 170 million, after launching a review-spam algorithm that examines longer-term patterns (Schwartz, 2024).

Yelp’s 2023 Trust & Safety report reads like a war diary: 278,600 user accounts closed for policy violations; 40,700 spam-associated business pages blocked; LLM-powered systems preventing 23,600 inappropriate reviews from publishing at all. “Defending the integrity and quality of content on Yelp is the foundation for everything we do,” said VP of User Operations Noorie Malik (Yelp Inc., 2024).

Why AI makes review floods different (and harder)

Pre-AI, fake reviews had tells: odd phrasing, repeated templates, strange timing. Generative models erased the accent. A 2024 peer-reviewed synthesis in The Knowledge Engineering Review warned that “counterfeit text” from large language models poses distinct detection problems because it’s both fluent and adaptable across domains (Adadi, 2024).

Early large-scale analysis in 2025 suggests AI-authored reviews often trade “specificity” for pleasing readability qualities lay readers sometimes rate as “high quality,” even as content becomes less diagnostic (Zhao, 2025).

Ad fraud researchers tracked the spillover: DoubleVerify’s Fraud Lab said “AI-powered fake reviews and testimonials” were seen across sectors and “grew more than 3x year-over-year” in 2024 (DoubleVerify, 2024).

The result: if you rely on human readers to sniff out fakes, you’re trusting the very cognitive bias the attackers exploit: fluency equals credibility.

Motives: extortion, sabotage, and pile-ons

Plenty of floods have a price tag attached. Extortion crews threaten a 1-star barrage unless you pay, often via gift cards because they’re untraceable and irreversible. The FTC’s consumer guidance is blunt: “Only scammers will tell you to buy a gift card … and give them the numbers off the back” (Federal Trade Commission, n.d.). The money lost to gift-card fraud is real and litigated. In 2024, Reuters reported a class action over Google Play gift-card scams was dismissed; the court noted the loss stemmed from scammers’ instructions to buy and relay card codes, reinforcing the “gift card = scam” baseline (Reuters, 2024).

Other floods are competitive or ideological. During intense geopolitical moments, businesses have been review-bombed far from the conflict zone. Coverage documented waves of “one-star ratings and threats” tied to the Israel–Gaza war that spilled into U.S. listings (Al Jazeera, 2023). Coordinated inauthentic behavior by “fans,” political activists, or bot-assisted networks often travels through TikTok/X/Telegram first, then detonates in your star rating.

And sometimes it’s a single viral video, someone tangentially connected to a brand, and the mob aims at the nearest listings, dragging unrelated employees and partners into the blast radius. There’s no proportionality in a feed.

If you felt alone navigating this two years ago, the policy frame changed. In August 2024, the Federal Trade Commission finalized a rule “banning fake reviews and testimonials,” empowering civil penalties and calling out AI-generated fakes by name. Chair Lina M. Khan’s language is unambiguous: “Fake reviews not only waste people’s time and money, but also pollute the marketplace and divert business away from honest competitors” (Federal Trade Commission, 2024). The rule bars the sale or purchase of reviews “by someone who does not exist, such as AI-generated fake reviews,” and prohibits paying for “positive or negative” reviews as such (Federal Trade Commission, 2024).

At the same time, major platforms formed cross-industry muscle. In October 2023, Amazon, Booking.com, Expedia Group, Glassdoor, Tripadvisor, and Trustpilot launched the Coalition for Trusted Reviews to share methods and “stop fake reviews at the source” (Tripadvisor/PR Newswire, 2023; AP News, 2023). It’s an implicit admission that single-platform defenses aren’t enough when review brokers and AI tools operate across everything.

Does it actually move revenue?

Short answer: yes - especially for categories where buyers can’t easily inspect the product or service beforehand (Handoyo et al., 2024). Empirical work in 2024 shows negative reviews measurably alter consumer search paths and purchase odds via clickstream shifts on large marketplaces (Varga et al., 2024). The scale of the manipulation problem also matters: an NBER working paper (2023) models how fake reviews shift beliefs most among uncertain consumers, precisely the prospects SMBs need to win (Akesson et al., 2023). Said differently: floods work because undecided buyers are, by definition, influenceable.

Why founders and SMB leaders should treat this like availability risk

Your brand’s online edge isn’t just copy or creative. It’s the ambient score attached to your name wherever customers decide. That score is as much “uptime” as any web service. Review floods aim to take it down long enough to redirect demand. The technical language from the DDoS world applies cleanly: adversaries coordinate distributed, low-cost inputs to overwhelm a public surface until legitimate traffic (real customers) can’t reliably access signal through noise (CISA, 2022; Cisco, 2025). In reputation space, the “packet” is a star plus 30 words. The attack looks different. The effect is the same.

The uncomfortable framing

Founders tend to look for a playbook. There isn’t one here - only conditions. AI text makes volume easy. Broker markets make access cheap. Social virality provides ignition. Platforms detect more every year, and still the flood finds the cracks. The FTC’s rule adds deterrence, but enforcement is always retrospective. Meanwhile, reputational DoS is a tactic anyone can rent.

If you run a clinic, a coffee chain, a SaaS with a Marketplace listing, a home-services startup in ten ZIP codes, this is operational reality, not PR folklore. As AP put it, “the internet is rife with fake reviews,” and the rise of generative tools is making the problem both faster and more convincing (AP News, 2025). That sentence lands like a shrug until the attack hits you on a payroll Friday. Then it reads like inventory.

References