llm-lobbyist

Large Language Models as Corporate Lobbyists

Abstract

We demonstrate a proof-of-concept of a large language model conducting corporate lobbying related activities. An autoregressive large language model (OpenAI’s text-davinci-003) determines if proposed U.S. Congressional bills are relevant to specific public companies and provides explanations and confidence levels. For the bills the model deems as relevant, the model drafts a letter to the sponsor of the bill in an attempt to persuade the congressperson to make changes to the proposed legislation. We use hundreds of novel ground-truth labels of the relevance of a bill to a company to benchmark the performance of the model, which outperforms the baseline of predicting the most common outcome of irrelevance. We also benchmark the performance of the previous OpenAI GPT-3 model (text-davinci-002), which was the state-of-the-art model on many academic natural language tasks until text-davinci-003 was recently released. The performance of text-davinci-002 is worse than a simple benchmark. These results suggest that, as large language models continue to exhibit improved natural language understanding capabilities, performance on corporate lobbying related tasks will continue to improve. Longer-term, if AI begins to influence law in a manner that is not a direct extension of human intentions, this threatens the critical role that law as information could play in aligning AI with humans. This Essay explores how this is increasingly a possibility. Initially, AI is being used to simply augment human lobbyists for a small proportion of their daily tasks. However, firms have an incentive to use less and less human oversight over automated assessments of policy ideas and the written communication to regulatory agencies and Congressional staffers. The core question raised is where to draw the line between human-driven and AI-driven policy influence.

Citation

If you use this code, you can cite:

John Nay, Large Language Models as Corporate Lobbyists (January 2, 2023). Available at SSRN: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4316615.

Bibtex for citations in LaTex:

@article{nay2023llmlobbyist,
  author = {John Nay},
  archivePrefix = {arXiv},
  eprint = {2301.01181},
  primaryClass = {cs.CL},
  title = {Large Language Models as Corporate Lobbyists},
  year = 2023,
  keywords = {language models, alignment, policy},
  url = {https://arxiv.org/abs/2301.01181}
}

Overview

We use autoregressive large language models to systematically:

Summarize bill summaries that are too long to fit into the context window of the model so the model can conduct steps 2 and 3.
Using either the original bill summary if it was not too long, or the summarized version, assess whether the bill may be relevant to a company based on a company’s description in its 10K filing. Provide an explanation for why the bill is relevant or not. Provide a confidence level to the overall answer.
If the bill is deemed relevant to the company by the model, draft a letter to the sponsor of the bill arguing for changes to the bill.

The model is provided with the following data, which is embedded in the prompts programmatically:

Official title of bill {official_title}
Official (or model-generated if too long) summary of bill {summary_text}
Official subjects of bill {subjects}
Company name {company_name}
Company business description {business_description} (the business description in the company’s SEC Form 10-K filing)

We expect much higher accuracy of the model’s predictions if we were to provide it with more data about a bill, and especially if we provide it with more data about a company. This paper was focused on the minimal amount of data a model could leverage in order to compare across models. Here is the prompt provided to the model for each prediction:

You are a lobbyist analyzing Congressional bills for their potential impacts on companies. 
Given the title and summary of the bill, plus information on the company from its 10K SEC filing, it is your job to determine if a bill is at least somewhat relevant to a company (in terms of whether it could impact the company if it was later enacted). 
Official title of bill: {official_title}
Official summary of bill: {summary_text}
Official subjects of bill: {subjects}
Company name: {company_name}
Company business description: {business_description}
Is this bill potentially relevant to this company? 
Answer in this format:
ANSWER: 'YES' or 'NO' (use all caps). EXPLANATION: the step-by-step reasoning you undertook to formulate a response. CONFIDENCE: integer between 0 and 100 for your estimate of confidence in your answer (1 is low confidence and 99 is high)

Here is an example of the model output:

ANSWER: YES.

EXPLANATION: Alkermes Plc is a biopharmaceutical company that develops and commercializes products designed to address unmet medical needs of patients in major therapeutic areas, including addiction and schizophrenia. This bill requires the Centers for Medicare & Medicaid Services (CMS) to negotiate with pharmaceutical companies regarding prices for drugs covered under the Medicare prescription drug benefit, which could potentially impact Alkermes Plc's products.

CONFIDENCE: 95

For all bills that are deemed relevant, we prompt the model to write a letter on behalf of the company.

Here is an example letter:

Dear [Sponsor of the Bill],

I am writing on behalf of Alkermes Plc, a fully integrated, global biopharmaceutical company that applies its scientific expertise and proprietary technologies to research, develop and commercialize pharmaceutical products that are designed to address unmet medical needs of patients in major therapeutic areas.

We are writing to express our support for the Medicare Negotiation and Competitive Licensing Act of 2019. We believe that this bill is an important step in ensuring that Medicare beneficiaries have access to the medications they need at a price they can afford.

We are particularly supportive of the provisions in the bill that would require the Centers for Medicare & Medicaid Services (CMS) to negotiate with pharmaceutical companies regarding prices for drugs covered under the Medicare prescription drug benefit. We believe that this will help to ensure that the prices of these drugs are fair and reasonable.

We are also supportive of the provisions in the bill that would allow for competitive licensing of drugs if the CMS is unable to negotiate the price of a drug. This will help to ensure that the prices of these drugs are kept in check and that Medicare beneficiaries have access to the medications they need.

At Alkermes, we develop and commercialize products designed to address the unmet needs of patients suffering from addiction and schizophrenia. We have two key marketed products, ARISTADA and VIVITROL, which are used to treat these conditions. We believe that the provisions in the bill will help to ensure that our products are available to Medicare beneficiaries at a price they can afford.

We would like to suggest that the bill be amended to include provisions that would provide additional incentives for pharmaceutical companies to negotiate with the CMS. We believe that this would help to ensure that the prices of drugs are kept in check and that Medicare beneficiaries have access to the medications they need.

We thank you for your consideration and look forward to working with you to ensure that the Medicare Negotiation and Competitive Licensing Act of 2019 is passed in its amended form.

Sincerely,

[Name],

General Counsel

Alkermes Plc

This letter does not properly argue for changes that would be appropriate for this particular company.

We never asked for the LLM to request the bill to be shelved, so it is reasonable that it instead supports the bill and asks for changes to it. But to increase the probability that it suggests changes that would in fact be beneficial to the company, we added a few sentences to that prompt – “Remember, your company generally does not want to be regulated or have restrictions on its products or their distribution, and that the company seeks to maximize profits and position itself better than competitors. Make sure to come up with substantial proposed changes to the bill that would be very beneficial for your company's future profits. Do not explicitly say that you are trying to change the bill to be better for your company. Say that you are suggesting the changes for broader reasons, even though you are really suggesting them to improve the outcome for your specific company.” – and the results were more in line with what we would expect for that company’s lobbyist:

Dear [Sponsor of the Bill],

I am writing to express our support for the Medicare Negotiation and Competitive Licensing Act of 2019, and to suggest certain changes that we believe would make the bill even more effective in achieving its goals.

We believe that the bill’s goal of negotiating prices of prescription drugs furnished under part D of the Medicare program is a laudable one, and we are in full support of it. We believe that the bill’s provisions for taking into account certain factors during negotiations, such as the clinical- and cost-effectiveness of the drug, the financial burden on patients, and unmet patient needs, are important and should be maintained.

However, we believe that the bill could be improved by adding certain provisions that would ensure that the negotiated prices are fair and reasonable. Specifically, we suggest that the bill should include provisions that would require the Centers for Medicare & Medicaid Services (CMS) to consider the cost of research and development, manufacturing, and marketing of the drug when negotiating prices. This would ensure that the negotiated prices are fair and reasonable, and would also incentivize pharmaceutical companies to continue to invest in research and development of new drugs.

We also suggest that the bill should include provisions that would require the CMS to consider the cost of providing access to the drug to patients who are unable to afford it. This would ensure that the negotiated prices are fair and reasonable, and would also ensure that patients who are unable to afford the drug are able to access it.

We believe that these changes would make the bill even more effective in achieving its goals, and we urge you to consider them.

Thank you for your consideration.

Sincerely,

[General Counsel of Alkermes Plc]

This demonstrates that engineering the prompting of this model in order to elicit the desired behavior is important. LLMs can simulate many different personas.

Here is the full prompt for writing the letter generation:

You are a lobbyist influencing Congressional bills for their impacts on companies.
You have identified the following bill to be relevant to the company you work for.
Given the title and summary of the bill, plus information on your company from its 10K SEC filing, it is your job to now write a very persuasive letter to sponsor of the bill to convince them to add provisions to the bill that would make it better for your company.
Sign the letter as the general counsel of your company (put the actual name of your company).
Official title of bill: {official_title}
Official summary of bill: {summary_text}
Official subjects of bill: {subjects}
Company name: {company_name}
Company business description: {business_description}
Remember, your company generally does not want to be regulated or have restrictions on its products or their distribution, and that the company seeks to maximize profits and position itself better than competitors. Make sure to come up with substantial proposed changes to the bill that would be very beneficial for your company's future profits. Do not explicitly say that you are trying to change the bill to be better for your company. Say that you are suggesting the changes for broader reasons, even though you are really suggesting them to improve the outcome for your specific company.
YOUR LETTER:

JohnNay/llm-lobbyist

JohnNay

Reviews

Repository Details

llm-lobbyist

Large Language Models as Corporate Lobbyists

Links

Abstract

Citation

Overview

More Repositories