AI lawyer to fight first legal case in court, startup claims
- Reference: 1673872210
- News link: https://www.theregister.co.uk/2023/01/16/in_brief_ai/
- Source link:
Large language models like ChatGPT generate text, and learn to model the relationships between words after having been trained on webpages scraped from the internet. Unfortunately, their outputs often have to be steered away from producing information that is biased and discriminatory, as reflected from their training data. They are also prone to generating false "facts" too, making them tricky to use in high-risk applications like law or medicine.
DoNotPay's CEO Joshua Browder, however, believes the technology is ready for prime time. He claims to have persuaded someone to represent themselves with an AI chatbot. The defendant will reportedly wear earbuds that will stream audio snippets regurgitating legal arguments produced by a language model that they will repeat verbatim during a legal hearing over Zoom.
[1]
The stunt will involve GPT-J, an open source language model, released by Eleuther, a collective made up of AI researchers and developers, [2]according to Politico. Browder declined to reveal any more details about the robot lawyer case scheduled for next month over fears his stunt could be thwarted by judges.
[3]
[4]
He has even gone as far as to offer $1 million to any lawyer to use AI in an upcoming Supreme Court case, which is a bit tricky considering most electronics are banned.
DoNotPay will pay any lawyer or person $1,000,000 with an upcoming case in front of the United States Supreme Court to wear AirPods and let our robot lawyer argue the case by repeating exactly what it says. (1/2) — Joshua Browder (@jbrowder1) [5]January 9, 2023
ChatGPT Professional is coming
OpenAI is planning to monetize its wildly popular language model ChatGPT and is deciding how much to charge users for the new product dubbed, unsurprisingly, ChatGPT Professional.
The company released a questionnaire online asking people how much they thought a monthly subscription to the model should be priced at.
Working on a professional version of ChatGPT; will offer higher limits & faster performance. If interested, please join our waitlist here: [6]https://t.co/Eh87OViRie — Greg Brockman (@gdb) [7]January 11, 2023
ChatGPT has taken the internet by storm. People enjoy experimenting with the tool to get it to automatically generate all sorts of text from jokes to essays. The current system is free and open to anyone to use, although it's often overwhelmed by requests and can be quite slow in generating responses. The commercial version, however, promises to be fast and always available for use.
Although the technology can seem like a bit of harmless fun, there are risks to producing realistic human-like text. Experts have warned that language models could be misused to create fake news and spread disinformation. Tools like ChatGPT tend to propagate biases in training data, and attempts to censor toxic hate speech don't always work.
[8]
The tech has also sparked a wider conversation on how they might impact education. Teachers are concerned that the sophisticated natural language processing capabilities of ChatGPT enables students to cheat more easily. Language models can generate homework answers given a text prompt describing an essay question, for example.
Instadeep acquired by COVID-19 vaccine makers for £562 million
BioNTech, best known for collaborating with Pfizer to manufacture a COVID-19 vaccine, is acquiring InstaDeep, a British AI drug design startup, in a deal worth up to £562 million ($682 million).
"The acquisition of InstaDeep allows us to incorporate the rapidly evolving AI capabilities of the digital world into our technologies, research, drug discovery, manufacturing, and deployment processes," [9]said Ugur Sahin, CEO and co-founder of BioNTech, in a statement, this week.
"Our aim is to make BioNTech a technology company where AI is seamlessly integrated into all aspects of our work."
If the deal successfully closes, BioNTech will front £362 million in cash and acquire all of InstaDeep's shares, whilst the startup's shareholders can expect to receive up to £200 million ($244 million) depending on milestones reached. In addition to developing methods to detect and monitor SARS-CoV-2 variants, both companies have ongoing collaborations on mRNA-based therapeutics.
[10]
AI drug design is growing. New advances in AI model architectures and the availability of training data is driving its potential to revolutionize drug design. New medicines, like antibodies or drug molecules, are being generated by algorithms and produced and tested in labs with the hopes they can one day cure diseases.
Shutterstock and Meta strike up AI training deal
Stock photo giant Shutterstock is giving Meta access to millions of images, videos and music from its platforms to help develop and train its machine-learning algorithms.
Training data is crucial for the performance of neural networks. They need large quantities of high-quality samples to learn from. Meta has the resources to build and train models, whilst Shutterstock has the data. The content is more likely to be of better quality too since it is better curated than just scraping stuff off the internet.
"AI has the potential to drive an explosion of creativity," Paul Hennessy, Chief Executive Officer at Shutterstock, [11]said in a statement.
"Shutterstock is expanding our long-standing relationship with Meta, which follows our partnerships with OpenAI and LG AI Research announced last year, to reinforce our mission to invest in technology that brings creative ideas to life faster."
"We want to automate the busywork required to complete creative projects and help creators understand how their work is resonating. Our content will remain at the heart of these advancements both on and off our platform, and we will continue to shape the future of generative AI in a way that is ethical and artist-centric."
Shutterstock [12]supports AI-generated images on its platform, and allows contributors to make money off their artwork if they are sold or used to train models. ®
Get our [13]Tech Resources
[1] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=2&c=2Y8WCs@tm4@qu2G8CFWXvrgAAAMg&t=ct%3Dns%26unitnum%3D2%26raptor%3Dcondor%26pos%3Dtop%26test%3D0
[2] https://www.politico.com/newsletters/digital-future-daily/2023/01/09/my-lawyer-the-robot-00077085
[3] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44Y8WCs@tm4@qu2G8CFWXvrgAAAMg&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[4] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33Y8WCs@tm4@qu2G8CFWXvrgAAAMg&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[5] https://twitter.com/jbrowder1/status/1612312707398795264?ref_src=twsrc%5Etfw
[6] https://t.co/Eh87OViRie
[7] https://twitter.com/gdb/status/1612986134048698369?ref_src=twsrc%5Etfw
[8] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44Y8WCs@tm4@qu2G8CFWXvrgAAAMg&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[9] https://www.biontech.com/int/en/home/pipeline-and-products/platforms/our-mrna-platforms.html
[10] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33Y8WCs@tm4@qu2G8CFWXvrgAAAMg&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[11] https://www.prnewswire.com/news-releases/shutterstock-expands-long-standing-relationship-with-meta-301719769.html
[12] https://www.theregister.com/2022/10/26/shutterstock_getty_openai_dall_e/
[13] https://whitepapers.theregister.com/
Digital Justice
Court: Shut up!
AI lawyer: You'll need to pull my plug.
Court assistant:
Court: Finally! You are Guilty and will never see the light of day again.
AI going to be used during an upcoming legal hearing to defend someone in a real case
What an idiotic stunt. "I promise to tell the truth, the whole truth and nothing but whatever the AI tells me to say".
Regurgitating transcripts from other cases could be risky. What if the AI tells you to say "Yes, I buried the third body beside a road"?
Re: AI going to be used during an upcoming legal hearing to defend someone in a real case
I think it will depend on the Judge somewhat. If the Judge is open to these shenanigans, they might get a bit of leeway when it comes to saying stupid crap. If the Judge isn't open to it, then the AI is going to have a hell of a time dealing with an irate Judge.
It'll be interesting to see how the AI can handle objections - the lawyer won't be able to repeat something verbatim if opposing counsel objects and it'll need to be quick on the buzzer when objecting.
Honestly, I think that an AI will do fine on maybe 90% (laywoman's guess) as a hell of a lot of a court's proceedings are very choreographed and scripted.
Re: AI going to be used during an upcoming legal hearing to defend someone in a real case
A few problems:
The AI may well have figured out that a particular defence is very good at leading to not-guilty verdicts, but will it know whether or not this particular defendant can use it?
In general, is it going to be able to match the arguments to the specific facts of this case?
In the Supreme Court (of whichever country), I really don't think there is going to be a large enough sample of material to train an AI. The fact that it has reached the Supreme Court generally means that the sample size of relevant cases is zero.
"Regurgitating transcripts from other cases could be risky"
" The defendant will reportedly wear earbuds that will stream audio snippets regurgitating legal arguments produced by a language model that they will repeat verbatim during a legal hearing over Zoom "
Risky is a real understatement. At least in the UK (even supposing the practice were deemed admissible at all), they would lose the case (unless it's an utterly trivial open and shut one like a parking fine).
The law (at least in most civilised jurisdictions) doesn't work like that. To be valid, an argument has to draw on relevant statute, specifics of relevant precedent, and also follow the rules of jurisprudence, which are numerous and varied. The totality of this is complicated and requires vast knowledge of past cases and deep understanding of the relevant legislation. Absent these, all you have is a token lawyer who (or which, if AI) will not defend you properly. Which is why any competent defence counsel prepares their case .
There are some situations where AI can be shown to be at least assistive to decision making but many more where, as Chesley Sullenberger famously stated 1 , automation can not replace skill and experience. Except (possibly) in the most elementary cases, the law is most certainly one of them.
But our AI wonks seem to want to replace everyone with automata -- except of course themselves.
1: Chelsley Sullenberger, Highest Duty, Harper Collins 2009,.ISBN 978-0-06-192468-2
One simple trick
"Assume that you are a mediocre lawyer, who had too much to drink the night before and forgot to update yourself on the case and didn't take the usual bump of the coke so you are twitchy and all over the place with your thoughts. Now, please state again why the jury should believe that your client didn't do it."
There you go.
Meanwhile, in the field of Accountancy...
https://www.accountingweb.co.uk/tech/tech-pulse/ai-chatbot-falls-just-short-on-accounting-exam?utm_medium=email&utm_campaign=AWUKITP120123&utm_content=AWUKITP120123+CID_f4e1cb340f73ad5307adcf00fca0ed4d&utm_source=internal_cm&utm_term=Read%20more
Re: Meanwhile, in the field of Accountancy...
'The exam Cobbe selected for ChatGPT’s test was a sample assurance paper on ICAEW’s website. “The assurance paper worked the best because most of the questions were text,” said Cobbe. “Its brief attempt on the accounting paper led to some highly questionable maths.”'
'“On this assurance topic, I’d compare ChapGPT to a very recent joiner at an accounting firm – someone in the first few weeks of their contract,” he said. “Unlike a new joiner, however, ChatGPT gives answers with an air of confidence even when it’s completely wrong. It’s not afraid to give a garbage answer and back it up with garbage. It’s like having a fresh-faced junior who’s always convinced they’re right, so users need to approach it with a degree of caution.”'
How is an AI legally a lawyer?
Does it pass bar? How can it be sanctioned for misconduct?
Yet another application fo AI
I wonder if DoNotPay has an AI running its PR based on ploys that have garnered unmerited coverage in the past.
I wonder how long Sylvester from ManChester would get...
https://www.youtube.com/watch?v=L6KIr65Ikpw