Summarization is a strength of generative AI, so we’ve tasked the tech with providing pithy summaries of earnings call transcripts.
Unless you have been living off the grid for the past year, you’ve encountered or used generative AI tools like ChatGPT, Bard, or others.
At Verity, we’ve made significant strides in incorporating intelligent, sensible GenAI features in VerityRMS. During this time, we’ve also interfaced with customers as we’ve pursued materially useful generative AI applications in our VerityData products.
As we and others have noted, generative AI has very real limitations in terms of what we can expect it to achieve successfully. GenAI’s hallucinations, or making up of information, is one of the main limitations — and a core barrier to investment teams. While GenAI is not yet reliable for providing new thoughts on any given topic, it excels at summarizing existing content.
In this blog, I will share the first GenAI enhancement released as part of VerityData | inFiling: summaries of earnings call transcripts.
Since summarization is a strong suit for GenAI, we’ve tasked the technology with providing pithy summaries of earnings call transcripts.
Verity’s GenAI Earnings Transcript Summaries are structured summaries that prioritize guidance, KPIs, and key drivers for companies reporting quarterly earnings. The summaries are generated from VerityData’s proprietary LLM and unique prompts guided by industry-leading investment researchers, resulting in a concise, factual, and unbiased summary for investors.
Based on what we’re hearing from customers and in the market, two use cases shine for the feature:
Customers can now subscribe to get these summaries for the companies on their watchlist as soon as the transcripts are made available. Interested? Please contact your customer success team to opt in.
The transcript summaries are one of many GenAI enhancements planned for the VerityData products. In the short term, we’re also working on AI-generated summaries of various sections of 10-Ks and 10-Qs, as well as period to period filings’ comparisons, while continuing to look at other applications.
In the long term, we’re especially excited about the possibilities for layering GenAI across a multitude of VerityData datasets — insider activity, management changes, buyback, institutional holdings, and more — that go back as far as 2004.
Our vision is to continue to enhance the mosaic of data points that investment teams need by triangulating:
While generative AI certainly has its limitations right now, its prowess in summarizing existing content can and should be considered seriously by investment teams. With this new feature, VerityData can provide teams with fast access to essential information, especially during high-volume periods like earnings season.
Accelerate analysis of SEC filings with clean, structured, and exclusive data available through VerityData | inFilings.