{"id":19115,"date":"2023-06-07T23:22:36","date_gmt":"2023-06-07T23:22:36","guid":{"rendered":"https:\/\/nftandcrypto-news.com\/crypto\/intuit-introduces-proprietary-large-language-models-for-fintech-with-genos\/"},"modified":"2023-06-07T23:22:38","modified_gmt":"2023-06-07T23:22:38","slug":"intuit-introduces-proprietary-large-language-models-for-fintech-with-genos","status":"publish","type":"post","link":"https:\/\/nftandcrypto-news.com\/crypto\/intuit-introduces-proprietary-large-language-models-for-fintech-with-genos\/","title":{"rendered":"Intuit introduces proprietary large language models for fintech with GenOS"},"content":{"rendered":"

<\/p>\n

\n

Fintech giant Intuit, whose product offerings include TurboTax, Mint, Credit Karma, Mailchimp and QuickBooks, recently expanded its software services platform to include GenOS, an operating system for generative artificial intelligence (AI) technologies.\u00a0<\/p>\n

According to Intuit, the new operating system will come with a suite of tools, including a developer studio, UX library, runtime layer and several pre-trained large language models (LLMs). <\/p>\n

\n

We\u2019re expanding our platform architecture to include a proprietary Generative AI operating system (GenOS) with custom-trained financial LLMs that specialize in solving financial challenges. <\/p>\n

GenOS will unleash the power of GenAI and ignite innovation at scale for customers.<\/p>\n

\u2014 Intuit (@Intuit) June 6, 2023<\/a><\/p><\/blockquote>\n

Several high-profile businesses have recently begun adapting third-party LLM solutions such as OpenAI\u2019s ChatGPT for their specific needs. However, Intuit\u2019s taken a different approach by creating proprietary tools and its own development and deployment platform. <\/p>\n

Intuit isn\u2019t necessarily known for its AI products, but its place as an industry leader leaves it well-positioned to leverage internal data to train models similar to ChatGPT. The primary benefit in doing so is that the company can cherry-pick what data gets included, thus allowing it to fine-tune its models for fintech. <\/p>\n

Where ChatGPT and similar LLMs, such as Google\u2019s Bard, have typically been positioned as general chatbots \u2014 meaning they\u2019re designed to discuss virtually any subject \u2014 a model trained specifically on financial data would be considered a \u201cnarrow,\u201d or targeted, system.<\/p>\n

And Intuit reportedly has a lot of data to work with. Per an announcement published on June 6:<\/p>\n

\u201cThe company has 400,000 customer and financial attributes per small business, as well as 55,000 tax and financial attributes per consumer, and connects with over 24,000 financial institutions. With more than 730 million AI-driven customer interactions per year, Intuit is generating 58 billion machine learning predictions per day.\u201d<\/p><\/blockquote>\n

It remains to be seen exactly how Intuit intends to implement GenOS, as the company hasn\u2019t so far disclosed any specific information about the LLMs it\u2019s currently developing through the new platform. However, some of the primary use cases for similar models have been in consumer education and customer service.<\/p>\n

Related: <\/em><\/strong>AI-related crypto returns rose up to 41% after ChatGPT launched<\/em><\/strong><\/p>\n

The launch of GenOS comes at a tumultuous time for United States taxpayers but could represent some relief for users of its flagship TurboTax product. <\/p>\n

The Internal Revenue Service (IRS) is currently under fire from U.S. conservative Republicans who\u2019ve proposed as much as $21 billion in budget cuts to the agency over the next two years. <\/p>\n

Such cuts stand to diminish IRS efforts toward modernizing citizen tax services, potentially negatively impacting an already complex tax filing system. This represents a problem that, combined with recent uncertainty surrounding the legal nature of digital assets in the wake of the Securities and Exchange Commission action\u00a0against cryptocurrency exchanges Binance and Coinbase, could pose significant challenges for the 43 million U.S. taxpayers\u00a0who hold crypto assets.<\/p>\n