AI Is the Primary Driver of Cybercrime Scale
An Interpol report finding that AI fuels more than half of cybercrime in Africa surfaced on HN, with the top comment being 'I'm surprised it's only half.' That deadpan reaction is the real data point: the community has already internalized that AI-enabled scams are the default, not the exception.
The thread connects to the broader TIME story also on HN today, where TIME is reportedly serving AI bots a different version of its website with ads built in. The incentives are misaligned everywhere: publishers are trying to monetize bot traffic, scammers are using AI to industrialize fraud, and the infrastructure running underneath all of it is largely the same open-weight models and fine-tuning pipelines that legitimate builders use.
The Bugtraq revival thread adds a wrinkle: the security disclosure community is noticing that Bugtraq's comeback announcement was itself AI-generated, which commenters found darkly ironic for a newsletter calling for renewed human connection in security research.
So what?
If you are building anything that involves identity, payments, or user-generated trust signals, your fraud baseline has shifted. The cost of running a convincing scam operation has collapsed. Founders in fintech, marketplaces, and communications need to budget for adversarial AI as a first-class threat model, not an edge case.