Do you need 700 engineers to make money with AI?
How to make impact with AI?
Recently, Builder.ai, a company founded in London and backed by Microsoft whose valuation relied on their revolutionary coding AI Natasha, was discovered to use 700 Indian engineers instead of an AI. Moreover, these engineers, it seems, often churned out rather buggy code. They are also accused of having inflated their revenue figures.
We are not here to tell you to hire engineers in India instead of working on your AI. Quite the opposite. I think while AI is not good for every use case and it needs work to be reliable, there are many opportunities, and you don’t need 700 people behind the scenes.
AI can be very useful. Last night, feeling overwhelmed by deadlines, I turned to my AI companion—within seconds, I could sketch out a plan for tackling and prioritising my tasks, we talked through my anxiety, and in the end, we even cracked a joke together. This morning, I asked AI to suggest a healthy diet plan from my fridge for breakfast, and to whip up a quick news update in my areas of interest—all before I’d finished my coffee.
These are not isolated examples; instead, they are evidence that AI is quietly taking over our everyday routines. From these everyday interactions, we start to see how user needs—like prioritising tasks, mental-health support, and on-demand information—inspire entrepreneurs to generate business plans built around generative AI. From therapists easing our anxieties to digital helpers managing our personal plans, generative AI is rapidly evolving from novelty to necessity.
In fact, if we view this as part of a much larger story, Marc Zao-Sanders, an AI industry analyst, released a report, based on a global survey of knowledge workers, showing that more than 68 % of knowledge workers now turn to generative AI for real-time problem solving—and nearly 40 % rely on it as an emotional sounding board.
These statistics not only highlight existing use but also reveal gaps that entrepreneurs can address. However, LLMs have their own problems that needs addressing through technical means. For example, LLMs still hallucinate facts around 15 % of the time and often lack deep domain expertise or the nuanced and broad context needed for professional outputs.
This underscores the need for technical expertise in developing solutions, in particular careful testing approaches, and robust monitoring. As a point in case, the recent survey by LangChain AI identified serious gaps in testing and evaluation.
Below are some of the top use cases from Marc’s research — each reflecting a specific user need that can be addressed by specialise existing LLMs including:
Therapy / Companionship
Enhance Learning
Generate or Improve Code
Generate Ideas / Brainstorming
Creativity
Healthier Living
A growing number of professional services are now being transitioned to generative AI—driven by the capability of LLMs to work with respective domains. While the performance of commercial LLMs is often already strong out of the box, many tweaks can be applied to go beyond and make it fit for professional use case that range from therapy and medical advice to legal counsel, tax guidance, personalised plans, and software development.
For an example of how generative AI has a potential to revolutionize an industry, if we look at AI in Mental Health, it was valued at roughly USD 1.13 billion in 2023 and is projected to grow at a 24.1 % CAGR through 2030, reaching about USD 5.08 billion—driven by rising demands for scalable, accessible mental-health support.
Woebot Health uses cognitive behavioural therapy via a friendly chat interface. Patients report a 30 % drop in self-reported anxiety after eight weeks of use. The company has raised roughly USD 114 million across eight rounds, backed by Temasek, JAZZ Venture Partners, and Leaps (by Bayer).
The advantages of AI therapist are clear:
It is available 24/7.
It is relatively cheap compared to traditional therapy.
It comes without the prospect of judgment from another human being.
These advantages are not just limited to the mental health sector, but generally span many different fields with specialisation process. It’s like having experts available 24/7 (with an internet connection) for the public. This is a game changer for many industries that couldn’t reach everyone due to cost and geographic resource limitations—for example, psychology or consultancy industries.
Despite these benefits, Woebot and similar AI therapists can sometimes offer oversimplified guidance, misinterpret nuanced emotional cues, and lack the personalised care of a professional therapist—so expert human oversight remains essential. A recent survey indicated that 62 % of users still prefer human therapists for complex emotional issues, highlighting the gap that specialised AI solutions must address.
This example demonstrates that introducing AI comes with complex challenges. AI is not magic and it can’t resolve everything. The big advantage of AI can help millions of users at once, whereas experts can allocate their time to only a few people a day. AI can serve many hidden needs that were previously unmet due to the difficulty of accessing specialised help.
Although there are already many companies in traditional industries that have started employing generative AI, there are still plenty of opportunities for entrepreneurs to spot unmet needs, work with LLMs to provide specialised services, and embed those services into everyday life. While AI still doesn’t match human ingenuity, they can help automate a lot of work that doesn’t need as much ingenuity.
I believe that the rewards are high for recognising demands and implementing innovative ideas. Although innovative ideas always carry uncertainty and risk, history shows time and again that those willing to take risks will grow and remain competitive.
Justin Ju
| AI Engineer at Chelsea AI Venture | justin@chelseaai.co.uk |
If you are interested in implementing AI solutions in your company, please subscribe to this channel. We will be sharing methodologies for implementing AI into business ideas and providing tech solutions to different problems.
At Chelsea AI Venture, we are offering services for scaling up AI in SMEs and we occasionally organise tech training events. Do not hesitate to contact me via email or by visiting our website for any questions or business cooperation opportunities.