Ask HN: If AI is good, why can't Gmail stop the following blatant spammer?
2 points by selfhoster 2 months ago | 2 commentsDoes Google not care people use their account system to spam people at other domains? If their AI is good, seems like they could filter these before they get sent, perhaps even detect abuse patterns of the abusers?
These all have a very similar format. I don't use SpamAssassin but I may have to go back to using it, to see if it can reject these emails.
-----snippet---------------
Dear Customer [redacted, my name],
We’ve received your payment of USD 357.25. Thank you for choosing Microsoft Office.
Your order has been processed successfully, and we’ve begun debiting the amount. It will be posted to your account within 24 hours.
To cancel your reservation, contact our support team at [redacted, likely the spammer's phone No.].
Please remember: Once placed, reservations are final and cannot be edited. We’re here to help with any concerns.
Subscription Details: Product Code: 7610634-460 Product/Service: InsightScan Security Services Email on File: [redacted, my email] Total Amount: USD 357.25
Best Regards, Microsoft Office Billing Team [redacted, likely the spammer's phone No.]
-----end snippet---------------
- cvhc 2 months ago(1) One would prefer precision over recall for spam detection, so bear some false negatives; (2) these classifiers are subject to adversarial attacks -- you may see spammers repeating the same tricks to circumvent the detector for a while and providers are slow to update their detectors.
- bell-cot 2 months agoTo paraphrase Charlie Munger - how could cutting down the spam improve Google's Quarterly Earnings Report?