PART -II
Disadvantages
We all know when Genie comes out of the bottle how dangerous it can be. ChatGPT too has BEEN out and needs to be approached with caution. A genie granting all wishes can be harmful. Even as we see remarkable capabilities, many questions and concerns are yet unanswered.
- While Chat GPT brings many advantages, opportunities and equal access,the technology also comes with certain risks and limitations as well. ChatGPT can also make errors and mistakes in its responses due to limited access to right and complete information and lack of verification. Given that Chatbots are trained on massive datasets, and its is possible that datasets may contain inaccuracies, inappropriate content and biases, chances are that bots may produce biased responses. These unsolicited responses may be harmful and confusing as well and can create an unhealthy perceptions and beliefs within individuals and groups impacting the environment.
- Neither the ChatGPT nor the individuals are completely trained on a perfect collaboration with each other. Chatbots may respond to one query or task at a time. The instructions or queries must be appropriately created and clearly articulated. The Chat GPT may get confused when multiple tasks and queries are raised and may fail to prioritise, as their intelligence as on today is limited (unless customised and developed further).
3.The other challenge with the bots as on day is that they may not be able to understand humour and sarcasm in queries or tasks generated by individuals as they are yet to perfect the human language. Take the following example. One air travellers’ luggage was mistakenly delivered to Hyderabad, while he had booked his ticket for Bhubneshwar. When he landed in Bhuvneshwar and realised the flight operators mistake, he immediately opened the chatbot of the operator and sarcastically mentioned in the chat box,” Thankyou for landing me in Bhubneshwar and my luggage in Hyderabad”. The bot instantly replied. “You are welcome. We will continue to serve you better”. ChatGPT may not yet understand feelings, emotions and can sometime provide irrelevant responses
- The overdependence on ChatGPT can also become a potential hazard. It can reduce the actual individual cognitive abilities and increase dependence on the technology where people may use it as a crutch to create, analyze and validate their tasks. Individuals run the risk of becoming hyper dependant on ChatGPT to such an extent where they risk to lose their originality and effort.
- There is a justifiable apprehension where academic integrity runs the risk of being getting compromised resulting into cheating and plagiarism. Researchers may replace their own efforts with ChatGPT and may use it for preparing their research proposals, literature review, data collection, analysis, producing research papers , thesis and publications. This substitution of learning effort can severely impair individuals own curiosity, originality, creativity, drive and initiative. The user needs to be educated about these risks and ethical dilemmas besides and also what are the far-reaching implications and how to balance the usage.
- We have seen in last few years how social media has created addiction and resulted into many problems, ChatGPT has a similar risk where students may develop unhealthy connections with the bots that may interfere their social and emotional development. There is further risk of the technology creating deep fakes and may negatively impact human interactions and relationships.
Conclusion
Perhaps avoiding and banning it seems unlikely, the teachers have to learn to develop strategies to use it as well as how to educate and communicate with students on its right, cautious and ethical usage and how it can damage or harm if not used appropriately. Students need to be taught on its nature and limitations. They need to be educated on its advantages and disadvantages and the methods to use it effectively to human advantage and how to balance its usage with other methods of learning including, group discussions, brainstorming, individual reflection and integrate with their own views and opinions. They need to be taught on difference between the different versions of Chat GPT (3.5, 4.0) and when to use it. It is important for humans to learn how we to stay two steps ahead of ChatGPT. Colleges and universities must also debate on what policies they should put in place regarding students using ChatGPT to complete academic work.
Humans in their endeavor for development need to bring a balance. There is a serious need to explore the eastern philosophies alongside western models. The ancient wisdom of Indian Rishi’s is a great tool for the humanity to stay relevant and create more sustainable technologies. The Artificial Intelligence ( AI) of contemporary times needs to be balanced with Adyatmik Intelligence (AI) of Bharat that is transcendental in nature. Its fusion is the need of the hour. In the recently concluded AI Safety Summit, held in UK, Prime Minister Rishi Sunak hosted Elon Musk and discussed the need to balance AI’s potential and risks.
However, despite its pros and cons, ChatGPT is here to stay, and from many market sources ChatGPT business is projected to grow and is expected to reach revenue of $1 billion by 2024. As per Elon Musk, “with AI, there was an 80 per cent likelihood of being net positive, and if we are cognizant about the 20 per cent downside risk, then, AI could be a great “equaliser”.
In a recent report of Best College Survey conducted amongst current graduates and undergraduates, 51% respondents believe that using artificial intelligence (AI) tools like ChatGPT to complete assignments and exams is cheating. Two in ten (20%) disagreed, and the remainder were neutral. Among those students who used AI tools to complete assignments or exams, 50% said they used them for only some portion of the work but completed the majority themselves. 30% used AI for most of their assignment, and 17% used AI to complete an assignment and turn it in with no edits.54% of the respondent students reported that their teachers had not openly discussed the use of AI tools like ChatGPT and only 25% said their schools or instructors had specified how to use AI ethically or responsibly.
With the rise in digital super intelligence, academic and research institutions, teachers need to be mindful of safety concerns and address them efficiently while educating themselves as well as the students. The governments also need to play the role of referee to ensure societies stay safe and educated.
In conclusion, Academic Institutions, Teachers, and Students must embrace the power of AI and ChatGPT, without forgetting the magic of human brain.
(Concluded…)
(The Author is Vice Chancellor, Shri Vishwakarma Skill University)