What is ChatGPT?
ChatGPT is an artificial language model developed by a company known as OpenAI, based on the GPT (Generative Pre-trained Transformer) technology. It is designed to generate human-like response writing based on the patterns and information it has learned from the training data to queries or prompts. Presently, it is in a question-and-answer format which makes it even more interesting, as if the questions are being answered by a human. ChatGPT has been trained on a massive amount of text data, allowing it to understand and generate reasoned responses to a wide range of topics. Its abilities include language translation, text completion, summarization, question answering, and more. This vast accessibility of Chat GPT comes with a vast database of knowledge feed to it.
However, with the unlimited possibilities that these applications can provide, some countries have felt threatened due to the lack of not having in place proper rules and regulations to govern the advancement of artificial intelligence. Therefore, many international forums and countries have attempted to take steps to protect their data privacy laws and have in place other laws to govern this artificial intelligence-based technologies.
What are The Global Trends and Recent Developments?
The Italian Government issued a report on ChatGPT and banned the use of ChatGPT in the country. However, they have recently lifted this ban. The major concern faced by the Italian government is that such AI based applications do not respect the data privacy laws of the country and there is a complete lack of transparency as to how the company stores and uses the data. This issue was also addressed by the European Union which later proposed various regulations for the same.
The UK government has a comparatively liberal approach towards the use of AI applications and has not proposed any restrictions. Although the United States of America does not have any federal laws for the use of AI, the country's National Institute of Science and Technology (NIST) does promote an AI Risk Management Framework. In India, the NITI AYOG and Planning Commission have issued guidelines on the working of such AI applications, however these are not legally binding. Countries like China, Russia, and North Korea with heavy internet censorships have altogether banned the use of ChatGPT in their countries.
Will The Consumer Forum in India Cover The Services Provided by ChatGPT in its Ambit?
The Consumer Protection Act, 2019 ("CPA"), protects consumers from any deficiency in service or any malfunctioning products. The major challenge is to determine if the user of AI / ChatGPT would fall within the definition of Consumer as provided under CPA. Section 2(42) of CPA defines a consumer, who is availing goods and services for a 'consideration' which means that the users of ChatGPT would be excluded from approaching the consumer forum since such users do not pay any fee to use ChatGPT. However, the upgraded version of ChatGPT does include paying of a subscription fee. Therefore, a user of ChatGPT could be considered as a consumer as per CPA and could approach the consumer forum. However, the disclaimers / terms of use of ChatGPT would prevent or come in way of the consumer from seeking reliefs. As per section 2(17) of CPA, Open AI would be considered to be a service provider under the electronic service provider definition and according to Section 2 (37) (ii) of CPA, Open AI would also be considered as a product seller as it includes a service provider. Would false information or a wrong factual matrix generated by AI amount to misleading the consumer or amount to deficiency in service? These questions still stand as roadblocks in determining the liability and accuracy of information provided.
Disclaimers and Terms of Use – Do They Provide Protection to ChatGPT?
The terms and conditions of use of AI applications are intended to reduce or mitigate the liability arising out of false content or responses generated by AI. Further, it defines the permitted conduct allowed by its user. A strong terms of use agreement is a great line of defence, but such defence is not absolute.
The important question that arises is how far these terms and conditions reduce the liability of the developers of AI applications? Would the terms and conditions be a sufficient defence if a user claims that the response generated was false / factually incorrect or misleading? A complete lack of clarity on such aspects in connection with the use of AI technologies along with its expansion bearing in mind, makes the framing of rules and regulations absolutely necessary.
Regulatory Framework – Need Of The Hour
Presently, there is no specific law governing ChatGPT in India. Undoubtedly, ChatGPT is a gamechanger, but there are many problems associated with its expansion. One of the major concerns with artificial intelligence is the uncertainty of the manner in which the algorithm works to generate a response and the criteria on which ChatGPT basis its replies on. If the algorithm has an inbuilt bias, it will undoubtedly lead to skewed responses. Does the answer lie in making the relevant algorithm transparent and available for scrutiny by the regulators?
Problems like accuracy of such information available, biasness of such information generated, source of such information, liability in case of relying upon such inaccurate / false information and many more issues are linked to the easy and accessible use of AIs. Further there is a lack of transparency regarding the usage and storing of data which is a major concern for data protection and privacy laws.
The recent technological developments have led to a new digital era which has opened the gates for AI development all around the world. ChatGPT is only one such application that has gained popularity and in a matter of no time there will be plenty of such applications open to public at large which may even be far advanced than what ChatGPT is today. Easy and speedy availability of information can be a major threat in the event of no stringent laws being in place imposing liabilities/responsibilities upon the developers of applications providing information through artificial intelligence technologies. Thus, a proper regulatory framework and stringent laws and policies must be enacted and is definitely the need of the hour. Even when such laws are in place, it would be an interesting journey for lawyers, judges and the policy makers as to how such laws would be interpreted to practically implement such laws.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.
