The recent launch of DeepSeek has generated significant excitement and headlines, and while the technology behind this latest iteration of Generative AI is undoubtedly impressive, its arrival also reflects the current state of AI – interesting, promising, and perhaps slightly overhyped.

This may be partly due to a generational difference. The baby boomer generation, who were among the first to work in IT, learned the lessons of business the hard way, with projects requiring robust ROI cases and careful alignment with business needs. In contrast, the current IT industry is vast, with major tech brands dominating the Fortune 50, and a focus on finding the next big thing. However, in the real world of corporate IT, a more cautious and pragmatic approach is necessary.

Artificial Intelligence (AI) is not new, with its roots dating back to the mid-20th century, yet it has only played a moderate role in the business world so far. The success of tools like Chat-GPT has brought AI into the mainstream, but it is still plagued by familiar issues, including high deployment costs, enormous compute power requirements, and questionable responses. Additionally, there are concerns about legal liability and copyright.

A Balancing Act

We need to strike a balance between the enthusiasm and experimentation surrounding AI and a healthy dose of pragmatism. We should start with the business case, asking how AI can help us achieve our mission and address strategic opportunities and risks. Currently, there is too much focus on “AI is great, let’s see what we can do with it” rather than a careful consideration of its applications.

AI presents a massive opportunity, but use cases need to be carefully worked out. AI excels at massive computation tasks, pattern recognition, and trend detection, and is surprisingly good at certain creative tasks such as image and music generation. However, it lacks the human sense of caution and experience of the real world, which is essential for making informed decisions.

AI is currently well-suited for tasks such as powering bots that answer predictable questions or automating rote tasks, freeing humans to focus on creative thinking. Robotic process automation has been a useful aid, changing the dynamic of human-computer interaction. Conversational AI is a work in progress, but we can expect rapid improvements as AI learns from iterative continuous learning.

However, we are still far from having robot CEOs, and AI “decisions” must be tempered by human judgment and common sense. The future will likely involve a collaboration between AI and humanity, but for now, we should deploy AI with care, sensible budgets, and a commitment to careful evaluation.

We should approach the next big AI breakthrough, such as DeepSeek, with a critical eye, querying its applicability, costs, and risks. As DeepSeek’s website says, “Into the Unknown” – we need to maintain a sense of adventure and optimism while avoiding getting lost in the new technological landscape.

Photo by Solen Feyissa on Unsplash

Want to learn more about AI and big data from industry leaders? Check out the AI & Big Data Expo events in Amsterdam, California, and London, which are co-located with other leading events, including the Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo.

Explore other upcoming enterprise technology events and webinars powered by TechForge here.