| Country | Canada |
| Pronouns | He/Him |
Brad Edwards
- 4 Followers
- 22 Following
- 19 Posts
Software developer, MSc Computer Science in AI student (Oct 2023), service dog handler, trail runner, former police officer.
@jwcph @emilymbender ChatGPT, Bard, Pi, etc haven't been marketed as truth engines. There's a million pounds of info out about how they're not and how LLMs work. For free. Everywhere online.
I'm just entirely unsympathetic to a professional who abdicates their responsibility by adding a tool to their practice without even a modicum of research into how it works. I don't see how that level of negligence bears on the chat bot providers.
@jwcph @emilymbender Sort of? Bing, you.com, Scite etc. They are explicitly positioning their tools as accurate within the bounds of what is accurate for their markets; search, academic databases, etc. And LLM systems can be engineered for that, primarily by using them only as summarization engines over standard search. I haven't seen any data comparing their accuracy to normal search.
After a certain point, "Don't touch the electric fence" means don't touch it.
The news stories on hallucinations, much less the research papers, industry coverage, etc are endless:
https://www.cspinet.org/blog/chatgpt-amazing-beware-its-hallucinations
https://www.nytimes.com/2023/05/01/business/ai-chatbots-hallucination.html
Law societies themselves have warned before this about problems with AI:
chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://www.lco-cdo.org/wp-content/uploads/2021/03/LCO-Govt-AI-Workshop-Report-%E2%80%94-March-2021.pdf
OpenAI ChatGPT:
"ChatGPT may produce inaccurate information about people, places, or facts. ChatGPT May 24 Version"
"GPT-4 still has many known limitations that we are working to address, such as social biases, hallucinations, and adversarial prompts. We encourage and facilitate transparency, user education, and wider AI literacy as society adopts these models. We also aim to expand the avenues of input people have in shaping our models."
@jztusk @emilymbender Bear in mind though, this wasn't a case of even using a search engine. That lawyer didn't use Bing chat, you.com or any other search engine that at least claims to provide results that exist somewhere.
The lawyer used ChatGPT. It's a make believe engine. And it's been widely advertised as such.
If that lawyer can't even bother to use an actual legal database, is the issue the tech platforms?
@emilymbender Other take. ChatGPT is /not/ Bing, Google, etc. And the lawyer wasn't using them. Those are marketed as search aids, and if they make up search results I think there's an argument to be made about not being fit for purpose. They've had to go to a lot of engineering to try and constrain LLMs to actual search results.
ChatGPT was explicitly not that. That lawyer is lucky he wasn't disabarred for incompetence.
@jwcph @emilymbender But they have said that. Repeatedly. And people have been yelling about that since ChatGPT was released. Repeatedly.
We can infantilize the public and put all of it on the companies, or hold people accountable for their mistakes. Disbar the lawyer, the others will learn right quick. Otherwise it's like banning Tide because some people thought eating Tide pods was fun.
@Magess @astatide @emilymbender They're not. That lawyer was simply foolish, misunderstood the product. ChatGPT was never marketed as a search engine. It's a deeply financed research project that went looking for a product later and was grossly misunderstood by the largely uninformed.
Companies are working on making LLMs useful coupled properly with search. There, the bet is a more natural interface with search will win more customers because it's easier to use and understand.
@Riedl Yikes. A lot of people do seem to struggle with the delta between 'erosion of privilege' and 'being discriminated against'. They're not the same and we really should be alive to people conflating them.