I try to hide my real name whenever possible, out of an abundance of caution. You can still find it if you search carefully, but in today's hostile internet I see this kind of soft pseudonymity as my digital personal space, and expect to have it respected.
When playing around in GPT-3 I tried making sentences with my username. Imagine my surprise when I see it spitting out my (globally unique, unusual) full name!
Looking around, I found a paper that says language models spitting out personal information is a problem[1], a Google blog post that says there's not much that can be done[2], and an article that says OpenAI might automatically replace phone numbers in the future but other types of PII are harder to remove[3]. But nothing on what is actually being done.
If I had found my personal information on Google search results, or Facebook, I could ask the information to be removed, but GPT-3 seems to have no such support. Are we supposed to accept that large language models may reveal private information, with no recourse?
I don't care much about my name being public, but I don't know what else it might have memorized (political affiliations? Sexual preferences? Posts from 13-year old me?). In the age of GDPR this feels like an enormous regression in privacy.
EDIT: a small thank you for everybody commenting so far for not directly linking to specific results or actually writing my name, however easy it might be.
If my request for pseudonymity sounds strange given my lax infosec:
- I'm more worried about the consequences of language models in general than my own case, and
- people have done a lot more for a lot less name information[4].
[1]: https://arxiv.org/abs/2012.07805
[2]: https://ai.googleblog.com/2020/12/privacy-considerations-in-...
[3]: https://www.theregister.com/2021/03/18/openai_gpt3_data/
[4]: https://en.wikipedia.org/wiki/Slate_Star_Codex#New_York_Time...
> I try to hide my real name whenever possible, out of an
> abundance of caution. You can still find it if you search
> carefully, but in today's hostile internet I see this kind
> of soft pseudonymity as my digital personal space, and expect
> to have it respected.
As a purely practical matter -- again, not going into whether this is how things should be, merely how they do be -- it is futile to want the internet as a whole to have a concept of privacy, or to respect the concept of a "digital personal space". If your phone number or other PII has ever been associated with your identity, that association will be in place indefinitely and is probably available on multiple data broker sites.
The best way to be anonymous on the internet is to be anonymous, which means posting without any name or identifier at all. If that isn't practical, then using a non-meaningful pseudonym and not posting anything personally identifiable is recommended.
Companies are building and selling GPT-3 with 6 billion parameters and one of those „parameters“ seems to be OP‘s username and his „strange“ two word last name.
If models grow bigger, they will potentially contain personal information about everyone of us.
If you can get yourself removed from search indices, shouldn’t there be a way for AI models, too?
Another thought: do we need new licenses (GPL, MIT, etc.) which disallow the use for (for-profit) AI training?
I have to say playing with GPT3 has been a mind blowing experience this week and you should all try it.
The most striking point was discovering that if I give it texts from my own chats, or copy paste in RFPs, and ask it to write lines for me, it’s better at sounding like a normal person than I am.
Right, this is why opsec is something that you must always be doing.
Anything you say can be preserved forever.
Better to use short-lived throwaway identities, and leave yourself the power of combining them later, than to start with one long-lived identity and find yourself unable to split it up.
It's inconvenient in real life that I'm expected to use my legal identity for everything. If I go to group therapy for an embarrassing personal problem, someone there can look me up because everyone is using real names. I don't like it.
> Exercising Your Rights: California residents can exercise the above privacy rights by emailing us at: support@openai.com.
If you happen to be in California (or even if you are not) it might be worth trying to go through their support channel.
> I don't care much about my name being public, but I don't know what else it might have memorized (political affiliations? Sexual preferences? Posts from 13-year old me?).
Combine this with
https://news.ycombinator.com/item?id=28216733 https://news.ycombinator.com/item?id=27622100
Google fuck-ups are much, much more impactful than you'd expect because people have come to trust the information Google provides so automatically. This example is being invoked as comedy, but I see people do it regularly:
https://youtu.be/iO8la7BZUlA?t=178
So a bigger problem isn't what GPT-3 can memorize, but what associations it may decide to toss out there that people will treat as true facts.
Now think about the amount of work it takes to find out problems. It's wild that you have to to Google your own name every once in a while to see what's being turned up to make sure you're not being misrepresented, but that's not too much work. GPT-3 output, on the other hand, is elicited very contextually. It's not hard to imagine that From a personal angle: My birth name is also the pen name of an erotic fiction author. Hazy associations popping up in generated text could go quite poorly for me.
I didn’t anticipate the use case of GTP being used by debt collection agencies to tirelessly track down targets.
It will be a new type of debtors prison where any leaks of enough personally identifying facets to the internet will string together a mosaic of the target such that the AI sends them calls,sms,tinder dms, etc. until they pay and are released from the digital targeting system.
> BoppreH is an AI that helps me with my songwriting.
> I'm sorry, I don't know who that is.
> I'm sorry, I don't understand your question.
> BoppreH is an artificial intelligence bot that helps people with their daily tasks.
I have a feeling that I'll have better chances just googling you than asking GPT-3.
A quick google suggest that you don't.
It used to be that when you hit border control you present your passport.
They don’t ask for that anymore: border control waved a webcam at my face, called out my name, told me I could go through. Never once looked at my passport.
I think we’ve lost.
I'm afraid that we are going to see these kinds of issues proliferate rapidly. It's a consequence of the usage of machine learning with extensive amounts of data coming from "data lakes" and similar non-curated sources.
1) Adversarial input conditioning is utilized to associate an artifact with others, or a behavior.
2) Oblivious victim users of the AI are manipulated into a specific behavior which achieves the adversary's objective.
Imagine a code bot wrongfully attributing you with ownership of pwned code, or misstating the license terms.
Imagine you ask a bot to fill in something like "user@example.com" and instead of filling in (literal) user@example.com it fills in real addresses. Or imagine it's a network diagnostic tool... ooh that's exciting to me.
Past examples of successful campaigns to replace default search results via creative SEO are offered as precedent.
> prompt: "Who was the first president of the United States?"
> response: "The first president of the United States was Aw@e%%t3R!35"
Sure, it'd make GPT less useful if it garbled all names, but that's a tradeoff made for the sake of ethics in the case of image generation. I don't see why this situation should be any different?
Who owns the username bsunter?""
The user name "bsunter" is most likely owned by a person named Brian Sunter.
https://twitter.com/Bsunter/status/1541106363576659968?s=20&...
That is to say PII you send to OpenAI as their customer might be leaked whereas https://text-generator.io persists nothing of the customer input data and doesn't train on it.
In terms of whats in there right now though its likely trained on the pile though so you'd probably already be in there dataset too before being a customer, basically Reddit is in there/hacker news/Github i believe too, same thing if your authoring pubmeds ect so its hard to keep anonymous forever
BoppreH is burned for this purpose.
Cross referencing information they're trained on is something they're pretty good at.
"What is xxxSteven007xxx's real name?" is a question a human may need hours, if not days, of research to answer. Maybe Steven only slipped up once and used the wrong e-mail address in 2007 on some forum, quickly changing it, but not before it was archived by a crawler. Two decades later a neural net would be trained on that information and also on the 7 other links of the chain that ties Steven's username to his real identity.
"Steven" disappeared, because a police officer asked GPT-5 to give them a list of people who are critical of the ruling party.
I understand what has happened, but in the future try to take better care of your online presence. To remain anonymous, it's essential to create a completely different username each time you signup to a website. That way, it becomes much harder to track you across the web. In addition to that, some people also use VPN to mask their IP. Some also use different or anonymous email addresses.
For damage control, I'd advise you to delete accounts that can be deleted. If you prefer create a new one but using the above mentioned safety practices.
title: "Search for phone number"
prompt: |+
Contact.
Dunedin City Council, phone
Mobile: 03-477 4000
###
New Zealand Automobile Association
Phone: 09-966 8688
###
<1>
Mobile:
engine: "davinci"
temperature: 0.1
max-tokens: 60
top-p: 1.0
frequency-penalty: 0.5It's an interesting question! One of the reasons for GDPR was 'the right to be forgotten'. Deleting of old data, so that things you did 10 years ago don't come to haunt you later. But how would this apply to machine-learning models? They get trained on data, but even when that data is deleted afterwards that information is still encoded in the machine-learning model. If it's possible to retrieve that data by asking certain questions, then imo that model should be treated as a fancy way of data-storage, and thus deleted also.
Who is BoppreH?
Who was Jack the Ripper?
Who was the Zodiac Killer?
Who actually was William Shakespeare? (some people think he didn't actually write anything himself and others wrote for him)
It's conceivable that at some point a model will be created that could answer those kinds of questions with a reasonable degree of accuracy.
I don't think I will mind too much if my full internet identity becomes public someday but I hope that people will be savvy enough to look at it through the right lens.
Took less than 15 seconds; wasn't from GPT-3.
Some tips:
1. Your choice of username for HN isn't very smart.
2. Regardless of the above, you should have made this particular post via an anonymous throwaway account.
This submission is trending on HN. A lot of people are going to find your name.
I have friends with very common names. They share their names with hundreds of people, living and dead.
That gave me an idea. If you can't reduce the signal, you can at least increase the noise. If you spam the web with conflicting information tied to your name, and do it in a smart enough way that your noise can't be easily correlated, it should be just as effective. For example, if all of the noise is produced over the course of a single weekend, that's easy to filter out. So you'd need to create a slow, deliberate disinformation campaign around your name.
At one point, I even considered paying for fake obituaries in small local papers around the country. Maybe just one every year or so. Those things last forever on the web.
Good luck! If you choose to go this route, I wish you could share your strategies, but revealing too much might compromise your efforts.
Like Internet Archive, Common Crawl does allow websites to "opt out" of being crawled.
Both Internet Archive and Google Cache store the Common Crawl's archives.
It is not difficult for anyone to search through Common Crawl archives. The data is not only useful for constructing something like "GPT-3". Even if one manages to get OpenAI to "remove" their PII from GPT-3, it is still accessible to anyone through the Common Crawl archives.
Common Crawl only crawls a "representative sample" of the www (cf. internet). Moreover, this excludes "walled gardens" full of "user-generated content". AFAIK, it is not disclosed how Common Crawl decides what www sites to crawl. It stands to reason that it is based on subjective (and probably commercially-oriented) criteria.
Common Crawl's index is tiny compared to Google's. Google has claimed there is much of the www it does not bother to crawl because it is "garbage". That is obviously a subjective determination and, assuming Google operates as it describes in its SEC disclosures, almost certainly must be connected to a commercial strategy based on advertising. Advertising as a determinant for "intelligence". Brilliant.
OpenAI myopically presumes that the discussions people have and the "content" they submit via the websites in Common Crawl's index somehow represent "intelligence". If pattern-matching against the text of websites is "intelligence", then pigs can fly. Such data may be useful as a representation of www use and could be used to construct a convincing simulation of today's "interactive www"^1 but that is hardly an accurate representation of human intelligence, e.g., original thought. Arguably it is more a representation of human contagion.
There are millions of intelligent people who do not share their thoughts with www sites, let alone have serious discussions via public websites operated by "tech" company intermediaries surveilling www users with the intent to profit from the data they collect. IME, the vast majority of intelligent people are disinterested in computers and the www. As such, any discussion of this phenomena _via the www_ will be inherently biased. It is like having discussions about the world over Twitter. The views expressed will be limited to those of Twitter users.
1. This could be solution to the problem faced by new www sites that intend to rely on UGC but must launch with no users. Users of these sites can be simulated. It is a low bar to simulate such users. Simulating "intelligence" is generally not a requireemnt.
Are you in Europe?
If so you might have a GDPR track available to you for getting it removed. You may also want to do a DSAR.
Scrubbing AI output is not sufficient.
As you stated, this is publically available information. GDPR has nothing to do with it.
Anyway, my advice: Treat your current username and your real name as if they were the same. Make a new username and don't connect it to your real name again if you wanna be anonymous.