Hacker Newsnew | past | comments | ask | show | jobs | submit | saberience's commentslogin

Paying ransomware fines is never the smart move to do unless you happen to trust what cyber criminals tell you.

You send them the payment, they tell you they deleted the data, but they also sell the data to 10 other customers over the dark-web.

Why would you ever trust people who are inherently trustworthy and who are trying to screw you? While also encouraging further ransomware crimes in the future.


It’s a sliding scale.

If you don’t pay, the odds they will publish your data are closer to 100%. If you do pay, the odds have historically been much closer to 0% than 100%

You aren’t paying to be sure, but to improve your chances.


I was guessing it's a OneDrive, Google Drive, DropBox or something.

Probably someone was phished and they still had access to an old shared drive which still had this data. Total guess but reading between the lines it could be something like this.


I guess it just means this: https://www.cybersecurity.ox.ac.uk/

"Cyber Security Oxford is a community of researchers and experts working under the umbrella of the University of Oxford’s Academic Centre of Excellence in Cyber Security Research (ACE-CSR)."


Probably, I'm not sure it's not https://gcscc.ox.ac.uk/

I don't think it's https://www.infosec.ox.ac.uk/

There's also this AI security research lab, https://lasr.plexal.com/

It looks like Oxford are quite busy in this space.


Passport or ID card scans would never be be stored alongside general KYB information, e.g. the standard forms PSPs use.

If you read between the lines of the verbiage here, it looks like a general archived dropbox of stuff like PDF documents which the onboarding team used.

Since GDPR etc, items like passports, driving license data etc, has been kept in far more secure areas that low-level staff (e.g. people doing merchant onboarding) won't have easy access to.

I could be wrong but I would be fairly surprised if JPGs of passports were kept alongside docx files of merchant onboarding questionnaires.


> Passport or ID card scans would never be be stored alongside general KYB information

How do you qualify this statement? Did you mean “should never”? Even then, you’re likely overstating things. Nothing prevents co-locating KYC/KYB information. On the contrary, most businesses conducting KYB are required to conduct UBO and they’re trained to combine them both. Register as a director/officer with any FSI in North America and you’ll see.


Fair point! Yeah, it could be. Although Europe tends to be stricter about those things, i.e. where PII is stored. I was trained way back in like 2018 about ensuring I never have any PII stored on my PC and around the requirements of the GDPR in terms of access to information and right to delete etc.

docx files of merchant onboarding questionnaires

Why would merchants fill out docx files? They would submit an online form with their business, director and UBO details, that data would be stored in the Checkout.com merchants database, and any supporting documents like passport scans would be stored in a cloud storage system, just like the one that got hacked.

If it was just some internal PDFs used by the onboarding team, probably they wouldn't make such a big announcement.


Another person wrote a good response to this but yeah, I would say, as someone that has worked in fintech, you will almost always have some integrations with systems which require Microsoft word format, as well as obviously PDFs, CSVs, etc.

Every country you operate in has different rules and regulations and you have to integrate with many third party systems as well as governmental entities etc, and sometimes you have to do really really technically backwards things.

Some integrations I remember were stuff like cron jobs sending CSV files via FTP which were automatically picked up.


If you are dealing with financial services (and payment provider most certainly would), you will be forced to interface with infuriating vendor vetting and onboarding questionnaire processes. The kinds that would make Franz Kafka blush, and CIA take notice for their enhanced interrogation techniques.

The sheer amount of effectively useless bingo sheets with highly detailed business (and process) information boggles the mind.

Some time ago I alluded to existence and proliferation of these questionnaires in another context: https://bostik.iki.fi/aivoituksia/random/crowdstrike-outage-...


That’s not how tax deductions work because a tax deduction doesn’t give you the full amount of your donation back it only reduces your taxable income, not your tax bill dollar-for-dollar.

Example:

You earn $100,000.

You donate $10,000 to a qualifying charity.

You can now deduct that $10,000, i.e. you’ll be taxed as if you earned $90,000, not $100,000.

If your marginal tax rate is 30%, you’ll save 30% of $10,000 = $3,000 in taxes. So you’re still out $7,000 in real money.


Though if that 100K to 90K move had actually changed your tax bracket, you'd stand to maybe save a bit more.

It changes nothing. If you get taxes 20% til 90k and 30% above that, then donating 10k still saves you 3k in taxes, you're still out 7k and you're still paying 18k in taxes on the 90k.

So, I used to work in the fintech world and it looks to me like what was hacked was merchant KYB documents. I.e. when a merchant signs up for a PSP they have to provide various documentation about the business so the PSP can underwrite the risk of taking on this business. I.e. some PSPs won't deal with porn companies or travel companies or companies from certain regions etc.

This sort of data is generally treated very differently to the actual PANs and payment information (which are highly encrypted using HSMs).

So it's obviously shitty to get hacked, but if it was just KYB (or KYC) type information, it's not harming any individuals. A lot of KYB information is public (depending on country).

Fair play on them for being open about this.


It's not just business data though - usually it will include ultimate beneficial owner and directors' passports, tax ID, etc. So there is a risk of identity theft there of potentially some very wealthy individuals.

Of course they wouldn’t announce acquisition and a license change at the same time but this is obviously the beginning of the end.

See Hashicorp and Elasticsearch for the same old story.

Luckily these kinds of products are a dime a dozen, ie zero technical complexity and there are so many similar projects already out there. Hell you can even vibe code this kind of project.


Yep embrace, extend, extinguish.

Ryadh from ClickHouse here, I commented below about the overall intent. Let me know if anything needs clarifying!

So you're ok with stealing the data yourself but not ok with providing it to others, ironic.

ChatGPT isn’t random though.

If you ask it what a star is, it’s never going to tell you it’s a giant piece of cheese floating in the the sky.

If you don’t believe me, try it, write a for loop which asks ChatGPT, what is a star (astronomy) exactly? Ask it 1000 times and then tell me how random it is versus how consistent it is.

The idea that non deterministic === random is totally deluded. It just means you cannot predict the exact tokens which will be produced but it doesn’t mean it’s random like a random number generator and it could be any thing.

If you ask what is Michael Jackson the entertainer famous for it’s going to tell you he’s famous for music and dancing. 1000/1000 times, is that random?


> If you ask it what a star is, it’s never going to tell you it’s a giant piece of cheese floating in the the sky.

Turn the Top-P and the temperature up. Turning up the Top-P will enable the LLM to actually produce such nonsense. Turning up the temperature will increase the chance that such nonsense is actually selected for the prediction (output).


Sure but nobody is doing that, are they?

I'm talking about the standard settings, and infact GPT-5 doesn't let you change the temperature anymore.

Also, that's not really the point. Humans can also produce nonsense if you torture them until they're talking nonsense, but that doesn't mean humans are "random."

LLMs are not random, they are non-deterministic, but the two words have different meanings.

Random means you cannot tell what is going to be produced at all, i.e. a random number generator.

But if you ask an LLM, is an Apple a fruit, answer yes or no only, the LLM is going to answer yes, 100% of the time. That isn't random.


I agree with everything that you've stated.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: