11 Comments
User's avatar
Vineet Mall's avatar

This was quite sharp and detailed. Haven't seen this depth across any other AI piece from India based folks.

Wish I had learnt about Activate sooner - would have defo pitched myself to join you :)

Vadiraj's avatar

Insightful and Useful stuff to gives a sense of the AI possibilities and use cases

Utsav's avatar

Very well put up. Let’s build for India

Harshith Viswanath's avatar

Hi! Interesting ideas. I think the two ideas that stood out to me are (1) AI-led Service Automation and (2) Startup Legal Automation. Service businesses have traditionally had very low margin and to increase revenue have led with a philosophy of hiring more people for more revenue. AI allows for revenue increase without hiring people. I think that a roll-up strategy could be employed by such startups wherein they acquire small service-businesses and optimize them with AI. This means that founders need to have a mix of skills such as Operations, M&A and embedding workflows. (2) Startup Legal Automation is a significant opportunity with the number of startups that are operating in India. I think an AI-Native Law firm for not only startups but also for MSMEs operating in sectors such as textiles, fisheries and agriculture. There are 5.93 crores MSMEs in India. BigLaw has traditionally avoided startups and MSMEs due to low-margins. However, these firms would embed AI as part of their core infrastructure thus capturing most of the value. The thing to note is that the CAC for these enterprises would be high but as the years go by these firms will handle more of their work (loyalty) generating high CLV. One of the other whitespaces for investment opportunities involves building Domain-Specific LLMs for specific practice areas such as M&A, IPR, Labor law etc. Founders have to go deep and understand one problem like patent research or a copilot for M&A transactions. This can be sold to law firms, in-house teams and AI-Native Law firms that are looking to embed AI tools into their workflows.

I write a blog titled "The LegalTech Thesis" wherein I analyze LegalTech trends and identify opportunities to build. Would love to connect with you and assist in identifying whitespaces for investment opportunities in the LegalTech sector.

Pratyush Choudhury's avatar

Harshith - excellent expansions, thanks for the depth!

Fully onboard with roll-ups in AI-led Services: acquiring fragmented low-margin players (BPO/ITeS, etc.) and injecting AI workflows is high-leverage - turns headcount scaling into margin expansion. Founder edge in Ops + M&A critical.

Spot-on scaling Startup Legal to MSMEs (6Cr+ opportunity in textiles/agri/fisheries). BigLaw's margin blindness leaves massive whitespace; AI-native firms can embed tools for value capture, trading high CAC for sticky CLV via repeat compliance/disputes.

Domain-specific LLMs (e.g., M&A copilot on proprietary corpus) add real moat - deep focus wins over generalists.

Love 'The LegalTech Thesis' - let me DM you to connect on whitespaces. Always hunting sharper LegalTech signals!

Cheers

Harshith Viswanath's avatar

Would love to help you hunt for more whitespaces in the LegalTech sector because I believe this is the perfect time for introducing technology in legal services! Will connect with you through DM.

rohan's avatar

good read.

Video AI required more support than any other stream. Text<image<audio<video

We need a good ecosystem(Startups/ Cloud provider / Foundational Model builder and Investors ) that understand that hierarchy and put the the actions accordingly.

Hard to crack but sweet.

Pratyush Choudhury's avatar

Rohan - thanks, sharp point!

Spot-on with the modality hierarchy: video's temporal + multimodal demands make it 10-100x harder than text/image/audio - compute, data coherence, everything compounds.

India's vernacular video boom (500M+ Shorts/Reels users, non-English surge) screams opportunity, but ecosystem lags: scarce GPUs, sparse Indic video datasets, risk-averse capital.

Hard to crack, absolutely - but sweetest upside in dubbing (regional OTT/edtech), editing (creator tools), generation (ads).

What video AI whitespaces are you most bullish on? Specific ideas or proxies you're tracking?

Appreciate the Signal!

rohan's avatar

Specifically in Editing (creator tools) - after shooting, content teams have to go through an editing layer which holds already shot video in inventory for 3-4 days. Creators don't have big content library unlike OTTs, so they want to post the shot video asap - editing layer act as a barrier for it.

This editing layer either need to be superfast/don't exist - we are building polarcut to solve this. Happy to chat sometime.

Rainbow Roxy's avatar

Hey, great read as always. I'm really curious if you see similar pivots towards value over sheer capacity happening as strongly in Europe, considering our diffrent talent pools and regulatory landscape? This focus on AI-led services and India's unique position is incredibly insightful, you truly have a knack for spotting where the next phase of the S-Curve is headed.

Pratyush Choudhury's avatar

Rainbow Roxy - thanks for the kind words and sharp question!

Europe's pivot to value (applications/services) is real but slower than India's—deep research talent (Mistral, etc) shines, yet AI Act/GDPR create friction: higher compliance opex, slower data flywheels, risk-averse enterprise adoption.

This results in stronger industrial verticals (manufacturing/auto AI with precision moats) vs India's services export velocity (cost + scale arbitrage). Less "sheer capacity" chase overall, more regulated premium plays.

India's lighter regs + talent density give us a deployment edge for global delivery.

What's your take on Europe's top whitespaces - specific verticals or pivots you're tracking?

Appreciate the insight!