How will Human-AI work relationships pan out? Expect career liquidity, jobs targeting specific problems rather than industries

Humans are still needed in the AI world. It is the how that will change, and the way we work that it will impact. (Image credit: Google Deepmind via Pexels)

The future as we know it is in the present. Observe and you will notice people living in their own versions of the future without their knowledge. Currently, I sit with a team of hard-core engineers and data scientists. They are building products with LLMs (large language models) and that is their present.

Honestly, before this, if someone spoke to me of artificial intelligence (AI) building blocks, I wouldn’t even pretend to know something about large language models, machine learning, large action models, AI agents or anything beyond Midjourney or ChatGPT.

Earlier, I prided myself on my AI ‘awareness’, as I knew the terms OpenAI, Sam Altman, generative AI and maybe shared the common belief ‘that AI is here as that large giant that will acquire all our knowledge and reduce us to lesser mortals who can no longer trust their own sensibilities’. Artificial intelligence was here to get us. That belief is now gradually evolving.

It will actually help to be less ambitious when starting out with AI transformation and take meaningful steps towards increasing velocity and throughput. (Image credit: Google DeepMind via Pexels)

‘AI is about small steps, not gigantic ones’

AI will be a steep change in organization velocity and productivity, no doubt. But that will happen through small steps, not moonshots. It will actually help to be less ambitious when starting out with AI transformation and take meaningful steps towards increasing velocity and throughput.

It will happen through strategic choices around which functions to automate with AI. From my conversations with founders building in the AI space, it is evident that the focus will first be on tasks where AI can add significant value even at relatively low-performance levels. Rather than magical applications, businesses will probably benefit from back-office work being automated first.

AI automation efforts will be perfect for areas where:

1. The value of automation is high (relaxes other high-value bottlenecks),

2. The consequences of failure are low,

3. The performance requirements will be achievable with ~20 points of IQ current
foundational models have.

Initially, the focus will be on tasks where AI can add significant value even at relatively low-performance levels. (Image credit: Google DeepMind via Pexels)

Use cases that ‘move the needle’ on business outcomes

Aakash Dharmadhikari, is the cofounder of realfast, a company building AI transformation for IT services. Earlier, Dharmadhikari held several senior product and engineering roles with Gojek, a super-app in Southeast Asia, including being its chief product officer. According to him, “AI will create ‘career liquidity’”- imagine you have a certain number of hard skills, AI can help you augment them by at least 5x.

Earlier, in the boomer-era, people had single careers for life, currently pivoting is the order of the day, soon, careers will be liquid. A single person can represent at least two-three sectors and multiple roles within the sector. Therefore, organizations might also have to think about how their roles are defined or may not need definitions at all (as long as tasks and outputs are well defined). ‘We are mostly heading towards humans using AI as a tool that can only help them do more.’ Further, one does not only have to be an engineer, as some of the hard skill sets can also be acquired/ learned.

Microsoft’s recent workforce survey (of over 31,000 people across 31 countries) shows that as of May, 75 percent of employees surveyed are using gen AI in their daily work — a figure that has nearly doubled since January — and 78 percent of those workers reported that they adopted these tools independently.

Nishant Singh is an AI enthusiast working with the global marketing team at TCS. As a marketer, he deals with a lot of content and data and AI tools have helped him simplify his work of social listening. In addition to that, Singh has been building his own AI products using the site There is an AI for That which helps find the right AI tool for the task at hand. The site helps you build videos, write SEO content, gives you content
ideas, summarize text, code, transcribe, design images and much more.

According to Singh, who took an AI education paid course, ‘working and building with AI is much more efficient than "studying" it per se’. AI is one of the only compute platforms that can speak to you and also nudge you towards asking the right questions. As per Singh and his building experience, prompting will become ubiquitous and all of us will eventually have to play around with it.

4C's of learning, and where AI tools will fit in.

Being ‘AI ready ’will matter for both individuals and institutions alike

Just like the internet has been an enabler and catalyst, of course with a double-edged sword, the same analogy can also be applied to AI. The only difference is that AI has proliferated much faster than expected and has taken the human mind some time to completely absorb and react to its impact.

iMature is an organization that works in the space of digital citizenship and internet maturity (DCIM). Its founder, Raghu Pandey, has been in the field of edtech for over 16 years, especially to make the internet a safe place for teenagers. He believes that the one reality we have to accept is that in the near future, we will have liquid careers and might have to be prepared to keep upskilling, and hence what will be important is
our capability to learn. Raghu and his team have developed a strategy to tackle exactly that — preparing for job liquidity and change. Watch the video here:

● Bottom: Internet maturity, skill development

● Middle: Understand market evolution (example: what is the next big thing to learn
online)

● Top: Preparing ourselves for the best-case scenarios presented by AI

The 4Cs are the basis of learning any of the skills — especially creativity is required to be a creative thinker (creativity is not only about art — it is about how you assess a certain problem). A framework to think about jobs is to design careers around problems and not industries, might continue to make one employable and coexist with AI.

Having said that, the struggle to adjust is real. The next four-five years might be excruciating across professions as they understand and adapt to AI. This is further propelled by the fact that colleges and universities are yet to include an AI curriculum despite the push by the National Education Policy 2020.

Overcoming these teething issues might just about change the order of work:

● JDs will have to be rewritten

● Humans might have their personal assistants

● The way certain tasks are performed might just be fully transformed and more…

We have to be prepared for a future where AI is an enhancement/augmentation tool to what we do and therefore the creativity/ challenge will lie in how we use it to our benefit. (Image credit: Google DeepMind via Pexels)

Humans and AI coexistence is inevitable and desirable. In fact, from what I have observed, it is the monotonous tasks that are being automated first, to be able to give humans more cerebral work. Therefore, to think that AI might just about wash us out might be premature and also incorrect. With AI, a lot of this tedious and repetitive work that's bottlenecking the value-creating work will be done away with. The conversation around AI and jobs often gets oversimplified into a binary of "jobs that will exist" vs "jobs that will be automated away". However, the discussion is complex, and this is certainly not the conclusion: "AI will make jobs disappear overnight".

The fact is this - we have to be prepared for a future where AI is an enhancement/augmentation tool to what we do and therefore the creativity/ challenge will lie in how we use it to our benefit as an individual and how we can train it and get it to do certain high-end tasks efficiently as a business. Looking at it this way, the opportunities have just opened up.

This does not mean that challenges will not crop up - people are already speaking of the double-edged sword - like data issues, confidentiality, trust and ethics - however, organizations like iMature and many others across the world are addressing these well within the grassroots by educating teenagers and students about the harmful effects of AI (just like they did with the internet). This issue will continue to be monitored and will continue to evolve as more people learn the rules of the game.