Using AI and ChatGPT? Costs and Implications for Businesses

A representational image of Artificial Intelligence (AI). Image is generated by Copilot AI.

In recent years, technological advancements have made artificial intelligence (AI) an accessible tool to both consumers and businesses. What are the most popular providers? Tech incumbents and early investors, including OpenAI's ChatGPT, Microsoft's Copilot and Google's Gemini. However, the advent of open-source Large Language Models (LLMs) has also lowered barriers to entry for AI development, allowing smaller companies and developers to build their own models. The result? Today’s AI market is crowded, with a diverse range of potential tools for businesses. However, firms should not focus solely on the technical aspects of AI. In our article below, we discuss how businesses can best leverage AI by considering its costs and wider implications.

Generative AI (Gen AI)

Generative AI is the technology’s most well-known variant, allowing for text prompts to create unique output such as graphics almost instantaneously. However, results can still be fallible. While Generative AI shines in terms of variety, convenience, and quantity, a qualified worker is often needed to oversee the final product. Still, newer models of Generative AI are increasingly capable of creating complex and specialised outputs, offering resource-strapped small businesses the benefits of increased productivity in more ways than one. In this way, Generative AI may be best placed to substitute missing human capital or relevant expertise. However, this leaves firms at risk of becoming overly dependent on AI technology providers. To prevent exploitation and to ensure that all businesses can move forward, industry leaders need to be aware of and mitigate these risks by forging mutually beneficial relationships.

Automation

In contrast to generative AI, agentic AI is a different class of AI technology which relies on largely self-sufficient decision-making systems to complete specific tasks. This allows businesses to leverage greater possibilities of automation, which similarly frees up valuable human capital for more complex and productive work. Yet, it should still be noted that human supervision is needed to refine output quality. On a micro level, firms which deeply understand the AI systems they implement are best placed to succeed. On a macro level however, a key concern is we have not yet seen the long-term effects of automation on this scale. For example, the opportunity to develop skills at work is an essential factor in honing young professionals into decision makers. In this sense, automation could limit the availability of crucial, industry-specific skills in workforce entrants. Forward-thinking corporations may thus need to consider how their actions today affect the next generation of human capital.

Processing Hardware

Most of the discussion around AI is focussed on software, which may cause hardware to become an overlooked aspect. In terms of technological development, improvements in computational power are also responsible for the modern proliferation and advancement of AI. However, manufacturers are increasingly struggling to keep up with the growing demand for affordable, accessible and computationally powerful hardware. This gives rise to a surge in prices and rapidly depleting stock. Demand concerns are exacerbated by the increasing popularity of cryptocurrencies and secure digital banking, which depend on much of the same hardware. For some, investing in the computational resources to utilise AI at an industrial scale may hold the answer to success over the next decade.

Sustainability

Despite the many benefits of AI, negative externalities such as environmental damage may also arise. Given the limitations of current hardware, generating each AI output significantly increases electricity consumption levels. The heavy usage of hardware also reduces its life expectancy and can potentially exacerbate the problem of e-waste. A short-sighted firm might dismiss these issues. However, experts know that when harmful problems arise, stringent regulation follows. Firms should remember: AI is just like any other technology. This means that in the medium to long-term, a narrow focus on efficiency and productivity considerations will not keep businesses looking forward. Organisations need the awareness and expertise to consider how ethics, responsibility and sustainability factor into their usage of AI. 

Beyond the Private Sector

So far, AI has seen widespread adoption by the private sector, however its possibilities in the public sector and in independent organisations should not be overlooked. With applications ranging from basic education to complex financial regulations, AI is poised to become a essential tool for all entities. 

In summary, AI is a powerful tool, but one which requires care. Businesses who use or invest in AI have a unique responsibility to keep up with legislative requirements, and to utilise AI in a way that brings long-term benefits. This entails an awareness of its risks and limitations, and the consideration of its impact on the wider world. Most importantly, we must remember that AI is one of many technological tools. No matter how advanced an instrument is, there is no substitute for good practice and ethical use.

****************************

Join JFourth's professional community here. Subscribe to our mailing list to receive curated insights on banking and compliance, business updates and ongoing networking opportunities.

Interested in seeing how AI can help with your firm’s compliance needs? Stay tuned for our next article on AI in Regtech, or discover JFourth Solutions’ Regtech services here.