India’s next big e-commerce challenge is not artificial intelligence itself but protecting millions of first-time online shoppers from AI-powered scams, according to Rohan Oommen, vice president for customer and partner trust at Amazon and the company’s global trust and safety chief. Oommen said consumer awareness has become as critical as technology in tackling digital fraud, as the rapid influx of new digital commerce users from tier-2 and tier-3 cities makes India distinct from more mature markets such as the US, where e-commerce adoption has largely plateaued.
Lower awareness, higher risk
"India has a large number of first-time users entering the digital economy. Awareness is much lower, which makes education a significant part of the solution," Oommen told TOI.
Generative AI has made fraud more sophisticated by enabling fake identities, forged documents and synthetic reviews, he said. It has also strengthened Amazon’s ability to detect abuse at scale, Oommen added.
The scale of fraud Amazon acted on
According to Amazon’s latest Trustworthy Shopping Experience Report, the company removed nearly 300 million fake reviews, took down about 70,000 phishing websites, and acted against around 14,000 scam-linked phone numbers globally last year.
| Fraud type | Global action taken last year |
|---|---|
| Fake reviews removed | Nearly 300 million |
| Phishing websites taken down | About 70,000 |
| Scam-linked phone numbers acted against | Around 14,000 |
Fighting AI with AI
Amazon is increasingly deploying AI systems, including agentic AI that can carry out continuous audits, while relying on years of behavioural data to distinguish genuine sellers from bad actors.
"Behaviour is much harder to fake than documents or images," Oommen said.
Attribution gap and platform accountability
Oommen said phone-based fraud remains harder to tackle than fake websites because establishing attribution is significantly more difficult. On platform accountability, he said Amazon does not wait for regulation before acting against fraud.
What this means for enterprise trust teams
For enterprises operating marketplaces or digital commerce platforms in emerging markets, Amazon’s approach illustrates the twin requirement the company identifies: AI-driven detection plus user education for first-time internet users. The company’s latest global figures — 300 million fake reviews removed, 70,000 phishing sites taken down and 14,000 scam-linked phone numbers acted against — show the scale of abuse that platform security teams can expect as synthetic identities and forged documents become easier to produce with generative AI. Amazon’s emphasis on behavioural signals, rather than document checks alone, points to the defence the company says is hardest for fraudsters to fake. The report’s numbers also show phone-based scams remain a persistent weak spot because attribution is harder to establish, tying back to Oommen’s point that consumer education must sit alongside technology.
Sources:
