Avaamo, a firm specialising in conversational interfaces using artificial intelligence, had recently raised $14.2 million in a round led by Intel Capital, with investment from Ericsson Ventures, Mahindra Partners, Wipro Ventures and WI Harper. Its platform allows companies to use omni-channel intelligent assistants in supply chain, HR, sales support, claims processing and insurance advisory. The firm plans to use the funds to expand its sales and marketing network. Avaamo co-founder Ram Menon talked about the product and the AI ecosystem in an interview.
How does natural language processing work in your chatbots used across countries?
Conversations are different and unique. Natural language is prone to colloquialisms, grammatical errors, accents and mispronunciation. [Our] approach is language-independent.
We can ‘teach’ the bot a new language in a few days. We already support pidgin languages like: Hinglish (Hindi in English characters), Banglish (Bengali in English characters) and Chinglish (Chinese in English characters).
Human interactions are often tainted by bias. Are AI systems capable of delivering unbiased service?
A recent customer feedback in the insurance industry was, the bot answers questions truthfully, and users prefer the bot over a salesperson because the bot does not ‘lie’ or make exaggerated claims but is factual.
Lack of bias can have a positive impact on the interaction.
How does your software compare with a human interface in terms of performance and cost?
Different customers have different motivations: some may look at cost reductions only, replacing humans; in other cases in India, people look at capacity expansion — can I handle 100% more insurance quotes during tax time without adding humans and still manage capacity; in still others, customers look at ways to expand their coverage across channels: voice, web, Facebook and WhatsApp.
Are investors queuing up to fund companies engaged in AI? Which other sector has got them excited?
It has quickly become clear that there is no monolithic AI market. AI is being applied by problem and by industry, and depending on the value proposition, investors are looking for real tangible benefits of the technology in terms of ‘making’ money or ‘saving’ money.
With AI being the buzzword of 2018, every company seems to be having something in the works. In what stage of evolution is the AI ecosystem and how many years before we see a shake-out?
The shake-out has already begun. Firms that are able to deliver tangible value and have a monetisation model that drives customer acquisition will succeed.
Avaamo has gained 70 enterprise customers in 18 months. They include Paysafe, Black and Decker in the U.S. In India, [we have] Reliance Capital, SBI Mutual Fund, City Union Bank, HDFC Loans and ICICI Prudential, Reliance Nippon Insurance, Axis Bank, Ashok Leyland, TAFE, Birla Sunlife, India First, among others.
Machines are learning to interact with humans. Will humans soon need training to interact with AI machines?
Not yet. There is still a lot of artificial in artificial intelligence.
In 2017, firms embraced digital payments. In 2018, they are focusing on AI. What does 2019 hold for us?
Blockchain for distributed ledger technology will redefine the way we think about ‘trust’ and is poised for a sea change in how it’s being implemented; managing privacy and user data is a huge problem that has not been solved well. This is an area which will see a lot of changes in the coming years.