Building in a world with Gen AI

We are at the cusp of an Industrial Revolution or so it seems with the rise of Gen AI. It has led to a frenzy of startups and tech big and small chasing the gold at the end of the rainbow. And yet, very soon we realized that while this technology is incredibly interesting, it is not enterprise ready yet.?

With its 100 billion parameters, GPT 3.5 can take you to the 80% answer, but getting to the 99% is hard.? Also the frontier of 99% is possibly not achievable for all classes of problems.?

This realization has not dampened the enthusiasm though; instead, it has prompted a more nuanced discussion about its future applications and the strategic decisions businesses need to make.?

Three key considerations if you’re building for impact and competitive advantage :

1. Solving for Context??

The ease of the 80% solution has dramatically lowered the barrier to entry in many fields, pushing startups to tackle more complex problems. For instance, the overnight obsolescence of many document processing startups built off Chatgpt highlights the shift away from narrow, point solutions.? If you have to go into panic mode every time Open AI releases a new impressive feature, then you have a problem.?

The market is increasingly favoring comprehensive, end-to-end workflows that address broader business needs, solving for specific outcomes. For example -? automating the process of writing cold emails for prospecting and lead generation to lift sales KPIs. If you are solving for specifics, Open AI and its foundation models getting better will only benefit you.??

2. Solving for Complex Interfaces?

Gen AI's ability to understand documents well represents another opportunity to simplify processes, especially in traditionally labor-intensive workflows requiring manual handoffs and validations.?

Now, there is an opportunity to reimagine how software is built, with AI at its core. And unlike traditional software, every component can be made intelligent and optimized, making the workflow better. This is where startups have a significant advantage, reimagining processes and building ground up AI led products.

Good design also plays a significant role in this. Building an intelligent system ground up will need to carefully define the role of a human in the loop. In fact , the more intelligent the system is , the more critical human feedback and explainability becomes because in the end AI doesn't have any OKRs and won't have a performance appraisal. If a human feels out of control with these intelligent systems then adoption is going to be low and there is a high risk of algorithms going haywire.?

3. Solving for Scale ?

Finally, Gen AI's potential to handle tasks at scale while minimizing manual drudgery presents a vast field of opportunities. Activities that are repetitive and time-consuming, such as data extraction, running queries, and data integration, are ripe for automation.?

However implementations can be expensive . Hence the effort vs ROI of AI implementations is justified when the value at stake is higher . By focusing on areas where high costs and manual processes are prevalent, you can develop solutions that not only improve efficiency but also allow businesses to reallocate human resources to more strategic tasks.

Imagine a world where Finance teams no longer spend endless days compiling financial reports. With AI assistance, this grunt work becomes a thing of the past, freeing up valuable time for CFOs to delve into market analysis and expansion strategies.

Our knack for creativity and strategic thinking is what makes Humans special. So before we hand over the reins to AI completely, let's make it tackle those tedious manual tasks first. There are plenty of inefficiencies out there for us to fix.

In conclusion, as we navigate the evolving landscape of Gen AI, it's clear that we're on the brink of something special. So while the technology may not yet be fully enterprise-ready, its potential is undeniable.

The key to success lies in embracing the possibilities while carefully navigating the complexities of implementation. By focusing on solving for context, complex interfaces, and scale, we can position ourselves at the forefront of innovation, driving sustainable growth and competitive advantage.