For a few days in February 2023, all people could talk about was a new artificial intelligence tool called ChatGPT. It’s a free-to-use chatbot that generates human-sounding responses from user prompts.
Techies are promoting it as a potential game changer for businesses and students, but many are concerned about its impact on the legal industry. As a result, some schools and universities have banned its use in the classroom.
Human in the driving chair
It is not easy to imagine an AI without a human in the driving seat. But it is possible to imagine chatGPT as a sort of human in the driving seat, one that takes its cues from the world and the people it encounters.
It also takes its cues from the human brain, which makes predictions and develops expectations of what is about to come next. This ability is called predictive coding, and it’s driving a lot of research in neuroscience at the moment.
Moreover, the ability to predict and develop expectations is part of what makes humans smarter than computers. It’s the reason we can pick up on other people’s emotions and reactions so quickly, even when they don’t tell us what they are.
Generative language models like ChatGPT use this same approach to predict what will happen next when a user interacts with them. They can learn to answer questions about topics ranging from weather to the history of the Roman Empire.
However, generative language models often have problems with the way they are designed and trained. These models are prone to sampling effects that lead to inaccurate outcomes, and they’re often compromised by social filters and other tacit assumptions.
While the technology behind these models is impressive, it’s important to remember that it is not the goal of these models to think like humans. It’s merely an instrument that can play with human knowledge and texts in a way that may be helpful, but it’s not a replacement for anything.
The technology behind generative language models can be an invaluable tool in helping people expand their perception of the world around them. In addition, it can help to increase their understanding of how their own behavior affects others.
This capability can be particularly useful for those who find themselves in a situation where they’re not sure what to do. For example, if you’re unsure how to write a good letter to a coworker or employer, chatGPT can help you create the best message that will get the response you want.
AI assists humans to expand human perception
One of the most exciting aspects of artificial intelligence is its ability to enhance human perception. AMP Robotics, for example, is using machine vision to sort recyclables out of trash. Others, such as Pensa Systems, use video cameras to improve shelf displays by removing clutter.
Another intriguing AI application, especially in a regulated environment, is the ability to spot and flag potentially fraudulent transactions. A number of companies, such as Stratify, have developed tools that can flag these equivocal items and notify the right people.
Among the most exciting developments are the advances in artificial neural networks that can mimic brain functions such as attention and learning. This is particularly important in a context like telecommunications, where a single mishap could lead to the ire of the customer or worse yet, an expensive lawsuit.
In the context of the real world, a more likely explanation is that we are on the verge of an era when a truly intelligent machine will be capable of accomplishing many tasks with equal or better results than humans at a fraction of the cost. Whether or not this will come to fruition is anyone’s guess. But one thing is for sure: AI will be a major player in the coming years. Those who take the time to learn about and explore these new technologies will be rewarded in spades.
AI knowledge true bicycle of the mind
A new chatbot, ChatGPT, by San Francisco-based OpenAI is gaining social media attention. Based on user prompts, it responds to questions with human-sounding responses that feel less artificial than earlier forays into AI.
The system has also become a talking point among academics and lawyers alike. The technology is expected to help answer legal questions, a task that attorneys have long struggled with.
However, AI isn’t without its limitations and ethical concerns. For instance, it can invent fictitious historical names and books, and it may fail to solve certain math problems. In addition, AI tools can raise privacy concerns.
There is a potential for AI to be used in the courtroom, which raises legal questions about how and whether it should be monitored. Moreover, if it uses language that violates personal rights, a judge might have to issue an order against the program.
This could be especially true if a judge believes that a text created by ChatGPT has prejudiced or discriminated against the person who provided it with instructions. Likewise, it may have produced false information or spread misinformation that infringes on a company’s reputation.
The use of AI in the courtroom would require a thorough understanding of the technology and how it works, says Zimprich. Nevertheless, he believes that legal firms should be willing to embrace chatbots for their potential value in helping with a variety of research tasks.
If a chatbot can perform more of these research tasks, it could free up attorney time to work on more important matters. In turn, their clients might be able to receive more effective service at lower costs.
It will be a challenge, but there are steps that can be taken to make ChatGPT more interpretable. This will allow attorneys to ensure that their chatbots don’t violate ethics rules and don’t give away confidential client data.
While ChatGPT is a fascinating new technology, it’s unlikely to be adopted across the legal industry for some time. Until AI is more transparent and its models are more interpretable, it will likely remain a niche tool for lawyers.
AI Bots can provide a Realtime Response
In Colombia, a judge has used an AI-powered chatbot to make a court decision in a case involving a dispute with a health insurance company. Judge Juan Manuel Padilla Garcia says he used the tool to assist him in solving a dispute between a health insurance company and the guardian of an autistic child. The case concerns whether the child is covered for medical treatment. The judge first posed legal questions to the chatbot and then used its responses as part of his judgement.
As a judge, Garcia was not required to use AI when ruling on the case but he did not want to deny its use. He said that he was using it to help him with jury selection matters and to brainstorm ideas for opening statements, but he also did not want to reveal the specific issue he had used ChatGPT for.
While judges are not prohibited from using AI in deciding legal cases, programs like ChatGPT often provide biased, discriminating, or incorrect answers. This is because language models do not actually understand the text they are translating. Instead, they synthesize words based on probability from the millions of examples they have been trained to read and respond to.
This is why a lot of people are not very comfortable with AI and it’s why some legal professionals are concerned about how it will impact the profession. Fortunately, there are some ways to ensure that chatbots and other forms of AI can be beneficial for the law industry.
Among these ways is by making chatbots and other forms of AI more transparent. This will allow users to interpret how the model comes up with its predictions.
In this way, a user will be able to draw conclusions about whether the model is being abused or not. This will be important because it can affect how the model is used in the future, as well as how it’s perceived by other people.
Additionally, it will be essential to make it possible for legal professionals to update the content on their firm’s website or blog with current data and statistics. This will help to ensure that a user does not get the wrong information or that their legal claims are in line with recent changes in laws.
Another way to make chatbots and other forms of AI more useful for the legal industry is by making them better at interpreting the language they are reading. This can help to improve how the AI responds to legal questions and it will also help to increase the accuracy of its predictions.