Talk-A-Bot

Jun 15, 2023

Talk-A-Bot Begins Integrating ChatGPT in Its Products Talk-A-Bot Begins Integrating ChatGPT in Its Products Budapest-based chatbot startup Talk-A-Bot began ChatGPT integration with the HR announcement-writing tool The integration was accelerated by the increasing number of internationa lblue collar workers coming to CEE and the growing need for interior multi-language communication The company insists that ChatGPT is only fir for corporate communication with thorough human supervision Hungarian corporate chatbot provider Talk-A-Bot announced the integration of ChatGPT into its HR announcement writing tool. It allows HR staff to increase their efficiency and overcome the language barrier. Four MBA students and CEU started Talk-A-Bot in 2016 as a homework project, which grew into Hungary’s first chatbot. The startup’s flagship is CHEQ – a chatbot for corporate communication. In contrast to most similar solutions that are meant for office workers, it has manufacturing industries and blue-collar workforce in mind and is available in 5 European and 5 Asian languages. Indelible Part of the Ongoing Product Development Earlier this year, Talk-A-Bot announced the conclusion of its a EUR 3.5M Series A round. According to the startup’s co-founder and CEO Ákos Deliága , about 2/3 of the funding is used to scale the international sales and marketing organization internationally while the rest of the funding is used to improve the overall product and develop features that are required in specific markets, such as the Notification Center for HRs. The two go hand in hand as specific markets require specific language pairs. ‘Central and Eastern Europe now needs foreign labor to keep our factories running. Every week thousands of new employees come to CEE from Asia. While we operate in Poland, Bulgaria and Hungary, we had to add Nepalese, Burmese, and Mongolian to the supported languages. We also started to work on integrating Kazakh and Uzbek languages,’ Mr Deliága tells ITKeyMedia. ChatGPT in Corporate Use Mr Deliága assures that when the EUR 3.5M deal with Széchenyi Funds was negotiated, ChatGPT wasn’t released yet. Moreover, back then, the startup insisted that OpenAI’s product wasn’t exactly fit for corporate use as it was prone to make critical mistakes. This includes the security concern. Olga Voloshyna, Chairperson of the Committee on IT and Cyber Security of the German-Ukrainian Chamber of Industry and Commerce, CEO at Silvery LLC Olga Voloshyna , the Chairperson of the Committee on IT and Cyber Security of the German-Ukrainian Chamber of Industry and Commerce , reminds of the recent errors and leaks relating to the corporate use of ChatGPT. ‘Latest technological advances can indeed be useful for work: when corporate chatbots employ GPT-4 with 170 trillion parameters, processed by Microsoft Azure’s supercomputers, they can solve complex problems at higher accuracy compared to proprietary corporate AI tools. But when companies choose to implement open source AI platforms, they must always assess the risks of confidential information getting leaked and remain extremely cautious of which information they are to share with such tools,’ the expert warns. ‘Upon thorough evaluation, we concluded that ChatGPT may impose certain risks for companies without human supervision. With human control ChatGPT becomes an incredibly powerful tool. And we will fully understand the real-life effective use cases only after the hype is over,’ Mr Deliága answers. +30% Efficiency with 95% Accuracy According to him, ChatGPT integration helps the HRs using CHEQ to speed up their work up to 30%. With the growing volumes of work in a company and the increasing diversity of its workers, the expectations toward HRs also grow as they de facto have to perform the work of a copywriter in several languages. CHEQ offers them assistance in the form of trustworthy, high-quality AI prompts. Ákos Deliága, Co-Founder and CEO at Talk-A-Bot Possible further inquiries will be processed by CHEQ’s chatbots. Talk-a-Bot claims that they are 95% accurate. As for the remaining 5%, Mr Deliága specifies the following three cases: The user’s question is too specific, and the AI’s knowledge base isn’t prepared to answer it. In such a case, there’s a possibility for human takeover. The AI misunderstands the question and responds with something irrelevant. That’s why it gives answers that indicate the question instead of a simple ‘yes.’ The user’s question is completely irrelevant and doesn’t warrant an answer. ‘The rapid advancements of generative AI are extremely inspiring. There are dozens of products that already improve our workflows and extend our capabilities within the company and also in our customer-facing products. ChatGPT is the most famous example at the moment, but the same technology is available for open source implementation. Eventually, the quality of the training data and the safeguards around the tool will tell which system is the best for a specific use case,’ the CEO sums up. The influx of international blue collar workers from Asian countries into CEE is among the less apparent consequences of the various factors of international turbulence. Language barrier in corporate communication is one of the arising challenges in this area. One can see how ChatpGPT arrived right on time to help Talk-A-Bot tackle this issue.