When Abhi Kumar based Voicing AI in 2024, his expertise with markets’ response to rising applied sciences got here into play, notably the way to take care of the preliminary resistance.

He had managed Microsoft’s rising market technique globally, co-heading its AI funding fund M12, and having deployed over $1.2 billion throughout 120 firms between 2019 and 2023. 

Again in 2014, when Satya Nadella took over Microsoft, Kumar remembers questioning the corporate’s management: what number of of you have got despatched a WhatsApp or WeChat message within the final 24 hours? Half the palms went up, largely amongst staff from India or China. 

“People had no thought about WhatsApp or WeChat. They used textual content messaging,” he stated. That perception ultimately formed Voicing AI. It was there he started forming a speculation: AI may play the identical function on this decade that browsers did within the Nineteen Nineties — an interface unlocking solely new enterprise fashions.

“When the primary browsers got here out, they didn’t maintain worth themselves, however linked individuals to the web and enabled PayPal, Amazon, Netflix. AI, to me, is the subsequent browser,” stated Kumar in an unique interplay with AIM.

The Overcrowded Agentic AI House

Voicing AI builds artificial brokers that may take calls, chat, reply to emails, suppose by means of buyer points, and full duties — basically changing the bottom tiers of name centre work (L0 and L1), whereas escalating complicated circumstances to people. 

By March 2024, Kumar stated, they have been passing the Turing take a look at in additional than 90% of circumstances. Nonetheless, know-how is simply half the story, the larger problem is determining the place AI matches into the messy actuality of enterprise operations. 

Kumar leans on his enterprise expertise to deal with this: “It’s not one of the best tech that wins, it’s the tech that’s applied to ship enterprise outcomes.”

Besides, the market is getting crowded. Increasingly more agentic AI startups are coming into the area. Gamers like Gupshup, RevRag.ai, Yellow.ai, Sarvam, and different AI startups have been racing to signal enterprise prospects.

It is because many of the agentic AI traction is within the B2B area and never for shopper apps. Startups are racing to stake a declare, however most are constructing on the identical underlying tech stacks, fine-tuning current LLMs quite than creating defensible IP.  

Learn: India has 109 Agentic AI Startups Constructing in a Vacuum

Considered one of Voicing AI’s rivals, Gupshup has been aggressively increasing its AI choices, shifting past messaging APIs to voice bots and buyer engagement. Founder-CEO Beerud Sheth had instructed AIM that he had additionally been getting ready for the agentic AI wave. 

Sheth believes that actual worth lies within the utility layer. That’s the place Gupshup’s current relationships with enterprises grow to be its greatest asset. The identical goes for Voicing AI.

In December final 12 months, LTIMindtree invested $6 million in Voicing AI for its proprietary know-how that brings human-like voice functionality throughout greater than 20 languages, whereas utilizing all the open supply instruments available in the market and fine-tuning these for particular use circumstances.

This demonstrates the significance of securing enterprise adoption over having a complicated mannequin. It’s why firms from messaging veterans to new AI-native entrants are crowding the identical subject, typically concentrating on the identical set of early enterprise adopters. 

In such an atmosphere, pace of implementation and integration into messy, real-world workflows turns into the true moat.

PoC Purgatory and Meagre Margins

Nonetheless, because the market crowds, the stress to win massive contracts — particularly in higher-margin geographies — will solely develop. Indian enterprise offers would possibly assist with credibility, however they gained’t drive the sort of income wanted to outpace world rivals.

Kumar believes the economics favour AI brokers. Within the West, labour prices are excessive, attrition in name centres is a continuing drawback, and multilingual protection is dear. AI can work 24/7, ship constant high quality, and swap between languages immediately. 

In concept, India must be the proper market: English is broadly spoken, the outsourcing trade is huge, and firms are underneath stress to chop prices. In observe, it’s the identical story that has performed out for SaaS, AI, and practically each different enterprise tech sector — Indian prospects love pilots, however not often pay.

Kumar finds Indian enterprises to be surprisingly “very very AI-first” in mindset, even when their budgets don’t at all times match their enthusiasm. “I really consider AI is the period the place India has an intrinsic benefit globally,” he stated, pointing to the hybrid human-in-the-loop mannequin that many firms will function underneath for years to return.

The economics, nonetheless, stay difficult. Within the US or Europe, changing or augmenting a $35,000-a-year customer support agent with an AI agent affords loads of margin to work with. 

In India, the place an equal function would possibly pay ₹3 lakh, the price financial savings are much less compelling, and a few firms nonetheless decide to “simply rent 5 engineers” quite than pay for automation, stated Kumar. 

This leads many Indian AI startups to look overseas for his or her highest-value prospects — a development that SaaS firms skilled a decade in the past.

The issue isn’t new. For over a decade, founders have grumbled concerning the “PoC purgatory” that plagues Indian enterprise gross sales. The sample is acquainted: a startup spends months constructing a pilot for a big firm, typically closely customised, however will get no contract, no income, only a “we’ll get again to you.”

Learn: Free PoCs are Killing Indian AI Startups

Kumar advises AI founders to suppose globally from day one, constructing for the markets the place the unit economics take advantage of sense.

He hasn’t written off India although. A few of Voicing AI’s largest prospects are right here, and he sees the market as a testing floor for operational complexity — a number of languages, excessive name volumes, and prospects who count on quick, personalised service. If a system can deal with that, it will possibly deal with nearly something.

The publish Voicing AI Loves India, However Is aware of Startups Nonetheless Wrestle Right here appeared first on Analytics India Journal.