How Venture Capital Could Use Large Language Models to Screen Sustainability Impact Startups
(2023) FMIM01 20231Environmental and Energy Systems Studies
- Abstract
- This study investigates the potential of large language models (LLMs), such as ChatGPT, to aid venture capitalists in the screening of startups that maximize sustainability impact. To determine the scope that maximizes impact for venture capitalists' and to identify effective screening criteria, the study utilized theoretical research and interviews. The thesis suggests that the ideal investment space is investments into high-risk, software-centric companies contributing to a sustainable system change that maximizes outcome impact instead of optimizing for environmental, social and governance metrics. This investment space along with other defined critical success factors were then deployed in an effort to test LLMs' efficacy in targeting... (More)
- This study investigates the potential of large language models (LLMs), such as ChatGPT, to aid venture capitalists in the screening of startups that maximize sustainability impact. To determine the scope that maximizes impact for venture capitalists' and to identify effective screening criteria, the study utilized theoretical research and interviews. The thesis suggests that the ideal investment space is investments into high-risk, software-centric companies contributing to a sustainable system change that maximizes outcome impact instead of optimizing for environmental, social and governance metrics. This investment space along with other defined critical success factors were then deployed in an effort to test LLMs' efficacy in targeting companies maximizing impact. Two prompting techniques were trialed, one question-based prompt where questions on critical startup success factors were asked, and another using a comparative method where the characteristics of screened startups were matched with investor profile preferences. In both versions of the model, the provision of context proved indispensable to analyze relevant startups, given GPT-4’s knowledge cut-off in 2021. Without context, the LLM often could not provide an answer or provided an imaginary one, especially for younger startups. The question-based prompting could accurately address some specific questions, while the investor profile prompt showed the most promising results by being able to efficiently summarize and present relevant output text on the given areas of interest. It was also found that the quality of the data input in the model directly affects its efficacy and it is therefore necessary to pick data carefully to avoid biases and greenwashing. This was especially true for question-based prompting, since the investor profile prompt was better at conducting an overall assessment of the companies with scarce information, but did still struggle to produce insightful ratings. In terms of the specific screening for impact startups, the model shows potential for targeting the ideal investment scope suggested by the thesis. The paper concludes by suggesting an immediate use case for the investor profile prompting technique in ChatGPT, supplemented by future use cases for automated systems to conduct outbound and inbound screening at scale. (Less)
- Popular Abstract
- How Venture Capital Can Unleash the Power of ChatGPT to Create a Sustainable Future
If you had 100 million dollars, how would you invest to create the most positive impact on society and the environment? Venture capital is frequently brought up as a driving force of systematic change in society and the investing form that can create the biggest impact. Yet, the integration of sustainability in venture capital has until recently largely been overlooked, only now are firms starting to recognize its potential. With the introduction of OpenAI’s revolutionary model ChatGPT, the things we thought were possible to achieve with artificial intelligence were propelled to another level. ChatGPT already has broad applications, spanning from code... (More) - How Venture Capital Can Unleash the Power of ChatGPT to Create a Sustainable Future
If you had 100 million dollars, how would you invest to create the most positive impact on society and the environment? Venture capital is frequently brought up as a driving force of systematic change in society and the investing form that can create the biggest impact. Yet, the integration of sustainability in venture capital has until recently largely been overlooked, only now are firms starting to recognize its potential. With the introduction of OpenAI’s revolutionary model ChatGPT, the things we thought were possible to achieve with artificial intelligence were propelled to another level. ChatGPT already has broad applications, spanning from code generation and essay writing to even aiding investment decisions. We propose that with the help of ChatGPT, venture capital firms can finally embrace sustainability as an integral part of their business model, helping them to identify startups that maximize impact.
Undoubtedly, investing in sustainable companies is paramount for solving the world’s social and environmental crisis. Venture capitalists’ (VCs) unique strategy, inclination towards short-term investments in high-growth ventures, has traditionally posed challenges for VCs incorporating impact investing. This is due to capital-intensive, long-term investments such as solar cells being on top of the sustainability agenda. By leveraging their existing capabilities to support high-risk, software-centric companies that drive sustainable system change, we argue that VCs can carve out a new, more appropriate space within impact investing. This novel and specific investing scope, requires a more sophisticated screening process than what is in place today. By harnessing the power of ChatGPT, our hypothesis was that VCs could refine and improve their way of identifying high-potential startups, and in turn find hidden gems in the impact space.
To test our hypothesis, we performed a study on ChatGPT’s ability to identify and differentiate the impact startups with the highest potential. By defining investor preferences (market, technology, founding team and sustainability) in a text, the model was then asked to rank the similarities of the investor profile and investment prospects. In addition, relevant company specific information was input in the model, derived from both the respective websites and publicly available data. The study shows that ChatGPT is able to reason intelligently about soft aspects related to the investment prospects in the VC space, especially when given additional information. The model struggles to perform comparable numerical ratings of the companies, but by summarizing key company information, it can quickly perform an insightful qualitative investment report. The model could therefore save a lot of time for screeners, while considering a larger amount of variables, like sustainability, in an early stage.
To perform optimally, the model requires extra data manually scraped from the internet and although inputting the data into the model is faster than reading it, the process is still time-consuming. Future iterations of the model should therefore be integrated into a larger software which scrapes the web for relevant company info, and automatically generates insightful reports on relevant startups. Thus, achieving an outbound VC screening of higher quality and greater quantity, that is more sophisticated in identifying the characteristics of a startup with the biggest potential for impact.
Although the current model has some weaknesses, scraping softwares along with rapid improvements of the technology, are likely to make LLMs an invaluable tool for VC investors in the foreseeable future. Allowing VCs to screen for the type of sustainable firms that fits into their specific strategy, and consequently help finance the impact stars of tomorrow.
This popular scientific article is derived from the master’s thesis “How Venture Capital Could Use Large Language Models to Screen Sustainability Impact Startups” authored by Karl-Gustav Elf and Måns Tivenius (2023). (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/9125296
- author
- Tivenius, Måns Vilhelm LU and Elf, Karl-Gustav
- supervisor
- organization
- course
- FMIM01 20231
- year
- 2023
- type
- H3 - Professional qualifications (4 Years - )
- subject
- keywords
- Large Language Models, Venture Capital, Impact Investing, Prompt Engineering, GPT-4, ChatGPT, Impact, Sustainability, Artificial Intelligence, Startup success, Impact startup, Impact measurement, Screening, AI for good
- report number
- ISRN LUTFD2/TFEM—23/5192--SE + (1-75)
- ISSN
- 1102-3651
- language
- English
- id
- 9125296
- date added to LUP
- 2023-06-15 10:46:56
- date last changed
- 2023-06-15 10:46:56
@misc{9125296,
abstract = {{This study investigates the potential of large language models (LLMs), such as ChatGPT, to aid venture capitalists in the screening of startups that maximize sustainability impact. To determine the scope that maximizes impact for venture capitalists' and to identify effective screening criteria, the study utilized theoretical research and interviews. The thesis suggests that the ideal investment space is investments into high-risk, software-centric companies contributing to a sustainable system change that maximizes outcome impact instead of optimizing for environmental, social and governance metrics. This investment space along with other defined critical success factors were then deployed in an effort to test LLMs' efficacy in targeting companies maximizing impact. Two prompting techniques were trialed, one question-based prompt where questions on critical startup success factors were asked, and another using a comparative method where the characteristics of screened startups were matched with investor profile preferences. In both versions of the model, the provision of context proved indispensable to analyze relevant startups, given GPT-4’s knowledge cut-off in 2021. Without context, the LLM often could not provide an answer or provided an imaginary one, especially for younger startups. The question-based prompting could accurately address some specific questions, while the investor profile prompt showed the most promising results by being able to efficiently summarize and present relevant output text on the given areas of interest. It was also found that the quality of the data input in the model directly affects its efficacy and it is therefore necessary to pick data carefully to avoid biases and greenwashing. This was especially true for question-based prompting, since the investor profile prompt was better at conducting an overall assessment of the companies with scarce information, but did still struggle to produce insightful ratings. In terms of the specific screening for impact startups, the model shows potential for targeting the ideal investment scope suggested by the thesis. The paper concludes by suggesting an immediate use case for the investor profile prompting technique in ChatGPT, supplemented by future use cases for automated systems to conduct outbound and inbound screening at scale.}},
author = {{Tivenius, Måns Vilhelm and Elf, Karl-Gustav}},
issn = {{1102-3651}},
language = {{eng}},
note = {{Student Paper}},
title = {{How Venture Capital Could Use Large Language Models to Screen Sustainability Impact Startups}},
year = {{2023}},
}