law .
Frederick : Speaking of that , some jurisdictions have claimed that any legal tech that generates forms for use by the public is considered the unauthorized practice of law . However , Georgia is not among them . ChatGPT is not holding itself out as providing legal services . In fact , I have seen disclaimers warning consumers not to rely on it for legal advice .
Neiditz : The use of a language model like ChatGPT may be considered a form of nonlawyer assistance . At a minimum , lawyers should establish guidelines and protocols for its use – and , when possible , contracts establishing direction and oversight by the attorney for uses involving separate non-attorney entities associated with the model . Attorneys should always review and approve or modify any decisions , conclusions or inferences made by the model .
Grossenbacher : Given the known issues with the accuracy of its responses , ChatGPT would not be well-suited as a proxy for preparing client or customer-facing legal advice . AI cannot replace the exercise of professional judgment when applying the law to the facts and providing legal advice . Using AI in this way would create a risk for malpractice claims if the information is inaccurate or otherwise flawed . AI can be leveraged by lawyers though . For example , using ChatGPT to create an employment policy is not so problematic . In practice , HR professionals often search for and obtain draft policies from the Internet and then revise them and have them reviewed by counsel before implementing them . Conceivably , an employment lawyer could do the same .
TAL : Another big concern with using ChatGPT in legal practice is confidentiality . Imagine that you are writing a brief in a case , feeding the machine all the necessary facts , and that information shows up when opposing counsel creates his document .
Neiditz : The answer is : Don ’ t do it . Disclosure to ChatGPT as currently configured may violate the duty of confidentiality , and the use of a language model like ChatGPT could lead to further disclosure of sensitive information if the model is not properly trained and secured .
To mitigate these risks , lawyers must ensure that one , the model is trained on a dataset that does not contain confidential client information , two , that any client information used as input to the model is properly anonymized , and three , that appropriate security measures are in place to protect client information .
With an enterprise version – offered by Microsoft and OpenAI - you can stop focusing on getting your people not to enter confidential information into large language models and start focusing on how to get them to enter permissible confidential information into the platforms that are strictly prohibited from sharing it with training data and ( if you qualify ) content reviewers . In addition , confidentialityrelated supervision should drive physical , technical and administrative security enhancements and protection through anonymization and firewalls that not only keep client confidential information out of the model but keep sensitive information in .
“ ChatGPT will make things up .”
TAL : Consequently , could disclosing confidential client information by a lawyer to ChatGPT jeopardize the attorney-client privilege ?
Neiditz : Yes . Again , the threat that ChatGPT poses to either client confidentiality or the attorney-client privilege depends on how the technology is used and configured .
TAL : And how well you train your employees .
Grossenbacher : Yes ! Allowing employees to use ChatGPT in connection with their jobs definitely impacts the confidentiality and data privacy of company information . As a language learning model , ChatGPT does have the ability to “ learn ” from every conversation . Currently , when a user logs into ChatGPT , a screen pops up stating that “[ c ] onversations may be reviewed by our AI trainers to improve our systems ” and advising users to “[ p ] lease [ not ] share any sensitive information in your conversations .” Thus , if employees are entering confidential employer information into ChatGPT , such information will clearly be used and shared with OpenAI to a certain degree .
IN THE PROFESSION
TAL : Which creates a liability for the employer .
Grossenbacher : Yes . You could even make the argument that an employer ’ s failure to prohibit employees from feeding ChatGPT confidential and proprietary information is not consistent with treating such information as a “ trade secret ,” and consequently , that information could lose its status as a trade secret . On the other hand , employees who use ChatGPT for their duties could unwittingly be exposed to the confidential information of others and use it in connection with their work . ChatGPT was trained on wide swaths of information from the Internet , and it is conceivable that it absorbed information that , while publicly available on the Internet , was trademarked , copyrighted and / or the intellectual property of another person or entity . If employees take possession of or use such information , this could create an additional legal risk for the employer .
TAL : In view of all these risks , what are safe and beneficial uses for AI in litigation ?
Neiditz : One of the main benefits of using ChatGPT is its ability to provide fast responses to routine legal inquiries and frequently asked questions . Private and enterprise versions of ChatGPT and its successors can also be programmed to provide customized responses based on specific client needs and preferences , making it easier for individuals and businesses to find the information they need . Another way that ChatGPT can make legal services more affordable and available is by automating repetitive tasks such as document drafting and data entry . It can help to streamline the process of preparing legal documents , reducing the time and effort required by attorneys and freeing up their time to focus on more complex tasks .
Wetherington : 100 percent of current litigation practices will be fundamentally changed by AI . Existing AI tools , like ChatGPT , can already replace large portions of " core " legal , marketing , and operational work for law firms . In the immediate future , the ethical , regulatory , and privacy concerns , now perceived as hurdles , will go away , and firms of all sizes will have access to low-cost AI to enhance document review , streamline legal research , draft documents , predict case outcomes ,
www . atlantabar . org THE ATLANTA LAWYER 15