ARTIFICIAL INTELLIGENCE
Where do we stand on AI developments?
As the hype settles down we need to take a balanced approach on the medium-term impacts of AI
Generally, as a revolutionary new technology washes over — there appear to be two clear categories of people. The cheerleaders who go gaga over it and cannot stop gushing over how it will change everything that we know about life today. And, the naysayers keep emphasizing that this is just a passing fad and that the impact won’t be significant. We are seeing this with AI.
To paraphrase, Amara’s Law — the phenomenon that we tend to overestimate the effect of a technology in the short run and underestimate the effect in the long run — is running its course on AI.
The near-term enthusiasm
In the past year, the landscape of artificial intelligence has shifted dramatically, marked by several key milestones that captured global attention. While several developments were underway, it was really the OpenAI’s launch of ChatGPT in November 2022 that quickly became a cultural phenomenon, reaching 100 million of users within the first few months and demonstrating the practical capabilities of generative AI in everyday applications. This milestone was rapidly followed by the launch of GPT-4, which further advanced these capabilities with more nuanced understanding and creative outputs. We started seeing stories, poems and this spurred a new sort of competition among tech companies in announcing their AI models and AI-integrated products. There was Bard which soon enough became Gemini then Meta’s models LLAMA, Perplexity, and Cluade, and even products like Humane’s Rabbit which sort of capitalized on the AI hype to position itself as the new age product replacing smartphone screens.
The Questions around Return on Investment
We are now sitting in mid-2024. While no one can deny that AI is indeed here to stay — the short-term enthusiasm seems to be mellowing down. The user retention and engagement is low, possibly because people are not finding enough value.
We are seeing reports on where the revenue will come from — Sequoia Capital recently published that given the amount of investments the AI companies need to deliver a return of $600 bn. Investors have added more than $2trn to the market value of the five big tech firms in the past year. Projecting an extra $300bn-400bn in annual revenues according to rough estimates, about the same as another Apple’s worth of sales, however, these returns are nowhere to be seen yet.
One big reason is while individuals are using AI chatbots regularly, enterprise use cases are not clear. The gains and near-term AI’s clear use cases seems to be in customer service. Klarna is one of the most touted examples — it recently claimed that AI was doing the job of 700 customer reps — however, speculations are that it could also be a claim to drum up press before going public. [Source: Economist] . The concerns around privacy, data security, hallucinations, and even the pace of change — i.e. companies don’t want to invest and then find out technology is updated and their implementation is obsolete are slowing down adoption. 2032 is now estimated to be the year by when AI is expected to start making a significant impact.
The Balanced Outlook
While AI has lofty expectations, a more grounded perspective is essential. It is clear that AI will undoubtedly reshape many aspects of our lives, however, its most effective role in the near to medium term is as a productivity enhancer rather than a wholesale replacement for human capabilities.
It excels at automating routine tasks, analyzing large datasets, and enhancing decision-making processes, which can significantly improve efficiency in various industries. However, expecting AI to solve all operational challenges is premature. The most effective approach is to use AI to complement human skills, allowing technology to handle repetitive tasks while humans focus on areas that require creativity and empathy.
Just like the personal computer changed the world in the late 20th century, artificial intelligence is shaping up to be the next big thing in technology. Bill Gates believes that AI will transform our lives just as deeply as computers and the internet did. He sees AI improving everything from how we learn in school to how doctors treat patients in poor countries.
Closing Thoughts
AI is here to stay for sure. Possibly the path from here will see a split of companies. Those building foundational AI models, akin to the “Intel Inside” model and those focusing on applications — which are able to show the product UI / UX needed for customer adoption. The real competitive advantage will shift from data model training to user experience design, as demonstrated by companies like Apple.
