[HN Gopher] Samsung workers made a major error by using ChatGPT
___________________________________________________________________
Samsung workers made a major error by using ChatGPT
Author : deesep
Score : 43 points
Date : 2023-04-05 17:46 UTC (5 hours ago)
(HTM) web link (www.techradar.com)
(TXT) w3m dump (www.techradar.com)
| withinrafael wrote:
| Confusing article. It appears the company discovered employees
| were pasting confidential information into ChatGPT and are
| assuming that data is now comprised given OpenAI policies stating
| conversations are periodically reviewed and used for training.
| The data doesn't appear to be accessible to the public directly.
| voytec wrote:
| It says at the beginning, that Samsung allowed employees to do
| so:
|
| >> The company allowed engineers at its semiconductor arm to
| use the AI writer to help fix problems with their source code.
|
| > The data doesn't appear to be accessible to the public
| directly.
|
| Allegedly (reddit), some random chats were recently listed in
| people's chat lists[1]
|
| [1] https://news.ycombinator.com/item?id=35236660
| navanchauhan wrote:
| Not allegedly, they actually were.[0]
|
| [0] https://openai.com/blog/march-20-chatgpt-outage
| withinrafael wrote:
| I forgot about that incident, good call! It's not
| inconceivable that one of these chats got reported back to
| Samsung.
| brundolf wrote:
| I think the market is going to explode (if it hasn't already) for
| on-prem, or at least private, LLMs on par with ChatGPT. This
| could be served by companies building their own, or by open-
| source projects, or by OpenAI or OpenAI's competitors
|
| As a side-effect, this feels like a bright spot in the
| potentially authoritarian trajectory that AI could take as labor
| becomes less and less valuable. It encourages development of LLMs
| that compete with the current default option and can be run on
| more and more limited hardware. Enterprises might even want
| separate departments, or separate individuals, to be able to run
| their own models to prevent leakage
| eachro wrote:
| Open source is a race to the bottom. Seems like the only
| obvious winner then is people selling the shovels aka NVIDIA.
| esafak wrote:
| More like a race to the top: you're forgetting all the
| applications that will be built on top of these open source
| models.
| eachro wrote:
| I'm not so sure. Does anyone pay for pytorch, numpy,
| tensorflow? In the matter of weeks we've seen llama.cpp,
| alpaca.cpp released to the public. Barrier to entry in this
| market is quickly going to zero.
| bob1029 wrote:
| How did they gain access to ChatGPT from their offices?
|
| I worked in the ATX factory about a decade ago and the network
| was _very_ locked-down at the time. You can 't even get your
| phone into the building without a security guard doing things to
| it. Taking basic stuff like paper in/out is also disallowed.
|
| I would have expected a total ban on personal computing devices
| leaving the parking lot if this happened during my time there.
| timcavel wrote:
| [dead]
| fullsend wrote:
| [dead]
| ftxbro wrote:
| The article says "now in the wild after being leaked" but then it
| says "the data is impossible to retrieve as it is now stored on
| the servers belonging to OpenAI." So did the source code leak out
| of OpenAI into the wild, or are they saying that OpenAI itself is
| "the wild"? As far as I see from the article, it's not accessible
| to the general public.
| brundolf wrote:
| If ChatGPT is trained on the data and ChatGPT is accessible to
| the general public, then the data may as well be accessible to
| the general public
| sterlind wrote:
| it seems unlikely to me that ChatGPT is directly trained on
| chat data. if it is, we should see it know information past
| its knowledge cutoff. afaik that hasn't happened.
|
| I assume the chat logs are instead training a reward model,
| which itself is then used as the reward function during RLHF
| training.
| bigyikes wrote:
| These models have a very long lead time before they're
| released to the public. Maybe GPT-5 is being trained on
| ChatGPT logs. I'm not sure we'd be able to detect if this
| was happening.
___________________________________________________________________
(page generated 2023-04-05 23:02 UTC)