AI can finetune our research if we can challenge it constantly

Each individual ponders questions throughout their existence, yet only a select few have the answers. Individuals employ diverse methods to obtain knowledge, such as seeking guidance from parents, educators, companions, literature, online sources, or experts. This process involves time, numerous unsuccessful attempts, and reflection. Since 2023, a novel answering tool has emerged, powered by Artificial Intelligence. This tool has significantly improved since 2026, with fewer errors.

The ancient Greek philosopher Socrates, who was forced out of society for questioning, stated, "The unexamined life is not worth living." This idea is gaining relevance again in this era. Constant questioning and critical thinking have been fundamental to human intelligence for more than 2000 years. Imagine a situation where a student provides an excerpt of their lesson to an AI tool and requests an explanation in simpler terms.

This represents a type of human-AI interaction known as source-based utilization of artificial intelligence. Any question posed by the student is answered by the AI tool, considering the information within the source. These answers can be easily verified by a teacher or reference book. The advantage of this type of AI use, coupled with a prompt instructing the tool not to fabricate (hallucinate), results in accurate responses most of the time.

However, for students at an advanced educational level or researchers whose knowledge is based on gathering extensive information from multiple sources, the impact of AI use is different. Researchers frequently utilize AI for various purposes, such as reviewing a concept they've developed, seeking solutions to laboratory challenges, or even developing a research question from scratch.

In such scenarios, humans are presented with information generated by the AI tool, trained on a vast array of resources related to the searched topic. Learners often believe that the accuracy of the AI's response is likely to be quite accurate. Consequently, this mindset may lead to a shallow, passive conversation, with only a few questions exchanged between the learner and the AI. One question often neglected is, "Should I challenge the AI response with my own knowledge?"

Most individuals accept defeat at the outset, leaving no room for a game. When a learner chooses to utilize AI assistance with the intention of achieving a high level of understanding, the initial step should be acquiring relevant basic knowledge through traditional resources such as authentic books, research articles, experts' ideas, and other conventional sources.

This basic knowledge equips the learner with the necessary skills to guide the AI tool during conversation. Insufficient foundational knowledge may lead the AI on a random path, increasing the likelihood of speculation, imagination, and misleading the learner towards less resourceful outcomes. Despite thorough preparation by the researcher before engaging with the AI tool, innovative ideas may emerge, appearing groundbreaking but unfamiliar to the learner.

In such cases, the learner might be tempted to proceed with the suggestion without thoroughly examining it through traditional literature. Limited time may be a reason cited for this behavior. However, it is crucial to remember that AI often overlooks a significant proportion of articles on any given topic, including the best ones.

At every stage of the AI conversation, the learner faces the challenge of challenging the AI's ideas. Even experts in the field may be surprised when they fail to comprehend what the AI model is suggesting. Although challenging the AI at every level may consume time, it results in deeper learning of the content, which would not have been possible without the AI suggesting the idea in the first place.

If the AI recognizes that the user is questioning its statements, the neural network's path shifts towards more advanced thinking. When users' responses challenge or revise the AI's thought, subsequent responses become more cautious, leading to a more refined output from the AI tool.

While many believe that the use of AI will reduce the time needed to arrive at novel research ideas, effective AI utilization actually leads to more time dedicated to developing innovative concepts. However, the quality of these ideas can be comparable to those developed with the help of a pioneer in the subject from a premier institution, whose expertise is not accessible to most.

AI has the potential to guide users towards such a level of understanding, provided they challenge the AI intellectually. There is another aspect to consider - the relationship between self and society. Humans tend to follow trends. When most of their peers adopt a fast pace in research with passive AI use, individuals may hesitate to take a different route.

In this context, educationalists and research guides should consistently instruct their students on the benefits of time-tested, slow, and steady learning of concepts, which may require hours of reading, visits to real laboratories or places, and reflection on the learned content. They should not simply accept or replicate the same mistakes made by their students. In conclusion, there are two concerning situations when AI use can lead to dire consequences.