Why AI Investment Announcements Do Not Always Lift Stocks

Initial developments in AI following the release of ChatGPT in 2022 have shocked the world with very quick technical and performance progress, moving from a chatbot surprisingly good at natural conversations to a system potentially able to replace a lot of white-collar jobs.

This shift in AI capacity has created a shift in the public discourse around AI technology as well. Enthusiasm about new capacity is increasingly mixed with worries about AI and is associated with responsible deployment, workforce displacement, or employment growth. And for sure, regular claims of mass job destruction by AI technology from developers of AI, like Elon Musk, have fired up these concerns, even when he tries to present it as a positive development.

“I think we are seeing the most disruptive force in history here. We will have something for the first time that is smarter than the smartest human. There will come a point where no job is needed. You can have a job if you want to have a job for personal satisfaction, but the AI will be able to do everything.”

As a result, AI announcements now function as both financial and reputational signals. A new study by two researchers at the University of Memphis and the University of Arkansas analyzed how AI announcements are seen by the market as either an innovation or a risk, and that the distinction is increasingly mediated by public sentiment.

They published their findings in The Journal of Strategic Information Systems¹, under the title “Do public opinions about ethics, layoffs, and hiring influence stock-market reactions to AI investment announcements? Insights from 2010-2023”.

Augmentation Or Replacement?

There are two potential paths when it comes to AI deployment in the workplace.

The first path is augmentation, where  IT products collaborate with humans, complementing each other’s capabilities, enabling mutual learning and enhancing each other’s capabilities. This is more in line with previous iterations of IT technology, and much less likely to meet significant opposition.

“Robotic Process Automation (based on rule-based engines) is an example of using AI for automation, while the use of speech recognition and conversational agents to improve first line customer interaction (based on Natural Language Processing) illustrates the use of AI for augmentation.”

The other path is replacement, where the AI system completely replaces jobs previously done by a human. In this context, employment in this position disappears entirely, and from a company point of view, salary costs are replaced by a (ideally smaller) new IT budget.

Augmentation is more suited to ambiguous tasks where AI systems complement unique human abilities, such as intuition and common-sense reasoning.

IT Investments And Market Returns

Previous studies have shown that IT investments impact stock returns and have a substantial effect on firm performance.

As a result, stock markets have generally “learned” to react positively to a firm investing in IT infrastructure. The same has initially been true with AI, as many companies announced the deployment of AI to boost productivity and growth.

Now, several years into this process, markets are starting to analyze the real effect of AI on productivity and employment with more discernment and analysis.

The growing concerns are not just about the effect on productivity, but also about ethical concerns. For example, AI could increase bias or increase societal inequalities by reducing employment and wages. The “black box” nature of many AI systems, where the decision-making process can be completely unexplained, only increases this risk.

“In the case of firms emphasizing AI investments in automation, AI system works independently and thus systems are left unchecked causing uncontrollable ethical issues. This may create negative perception about AI. Thus, opinion about ethics arising from the adoption of AI in automation could diminish the perceived potential benefits from AI-enabled automation”

AI Announcements & Stock Markets

Assessing AI Investments

The study analyzed announcements made by U.S. publicly traded firms on AI from 2010-2023, using LexisNexis data.

Out of the 1,006 announcements, the researchers excluded the ones too likely to be tied to other effects, such as when a merger, acquisition, divestiture, or change in CEO occurred in a 14-trading-day event window around the AI investment announcement. In the end, they analyzed the market returns created by 403 unique AI announcements by U.S. publicly traded companies.

These announcements were later analyzed for their content related to topics related to employment likely to affect perception of AI.

“For each individual tweet by investors, we used a Python program to extract the sentences mentioning ethics, layoffs, and hiring related keywords. Throughout the paper, “opinion about ethics,” “opinion about layoffs,” and “opinion about hiring” refer to the tone of the discourse on each topic in the tweets surrounding an announcement.”

The study then looked at financial return following the announcement about AI implementation. The focus was not on the market’s immediate reactions, but the effect on the stock price several months later.

“We use the one-year buy-and-hold abnormal returns (BHAR) from the day on which the AI investment is publicly announced.”

Impact Of AI On Markets

The first discovery of this study is that for the time period 2010-2016, the mean compounded daily returns between AI and non-AI IT announcements do not differ significantly. This makes sense as during this period, AI was in its early stages and somewhat similar to other IT systems used to augment the existing workforce.

In contrast, for the time period 2017-2023, the mean compounded daily returns for AI firms significantly exceed those for non-AI IT firms. This reflects the increasingly positive view of investors about AI and its future effect on productivity and a specific stock’s prospects.

The study found that automation emphasis, augmentation emphasis, and positive public discussion about hiring were associated with stronger long-term abnormal returns. By contrast, more negative discourse about ethics and layoffs was associated with weaker returns.

“Emphasis on both automation and augmentation, and opinion about hiring, are positively associated with BHAR, both opinion about ethics and opinion about layoffs are negatively associated with BHAR.”

Source: The Journal of Strategic Information Systems

Investing Takeaways

AI is often anthropomorphized, meaning people attribute human characteristics, intentions, or capabilities to it. This is both an advantage, as it can definitely do tasks that other IT innovations would be unable to achieve, but also a risk, as it is increasingly perceived as a competition to actual humans, and not just a force-multiplier for workers.

For this reason, many associate AI with ethical concerns and fears of job displacement.

So the association of the effects of AI alongside AI investments can change the perception of the market and the result on stock prices.

Both automation-focused and augmentation-focused AI announcements were positively associated with long-term abnormal returns. However, those relationships changed with the surrounding public discourse, particularly when automation was discussed alongside ethical concerns, layoffs, or hiring.

“When AI investment emphasizes automation, the positive association with long-term abnormal returns weakens as public opinion about ethics and layoffs becomes more negative but strengthens as public opinion about hiring becomes more positive”

One explanation for public perception is that markets might not distinguish layoffs due to AI improving efficiency from layoffs due to financial struggle. And that reaction might be partially warranted, as falsely blaming downsizing a company’s workforce on AI can be tempting for any company experiencing a downturn.

And of course, AI investments are not guaranteed to turn into positive returns on the invested effort and capital.

So with AI announcements, investors are evaluating not only the technology’s potential, but also implementation credibility, labour consequences, regulatory exposure, and whether the investment has a plausible route to revenue or efficiency gains.

The study does not suggest that simply adding “AI” to a corporate strategy guarantees a valuation premium. Instead, it shows that long-term market outcomes depend partly on what the investment is intended to accomplish and the public discourse surrounding its workforce and ethical implications. Because the dataset ends in 2023, whether investors have become still more selective during the subsequent generative and agentic AI investment cycle remains an important open question.

Investing in AI Applications

RELX PLC

Deploying AI does not guarantee a return. The strongest commercial cases connect AI spending to proprietary information, established customer relationships, and products with a measurable path to monetization.

A good example of such a database useful for AI is RELX, a global provider of information-based analytics and decision tools for professional and business customers. Its 37,000 employees serve customers in 180+ countries. The company was previously called Reed Elsevier and is a major scientific publication publisher. During the 1990s- 2010s, the company made a long list of acquisitions to expand its data business in supply chain, online IDs, regulatory risks, anti-money laundering, law enforcement, etc.

The core business of the company is split into 4 different segments.

The first one is risk, using public and industry-specific content with advanced technology and algorithms to assess and predict risk.

Source: RELX

Another segment is the historical center of the company’s publishing: scientific, technical & medical data and publications, of which the most prestigious publication is likely the medical journal The Lancet.

The third segment is legal data and advice, through its LexisNexis Legal & Professional branch.

“We help lawyers win cases, manage their work more efficiently, serve their clients better, and grow their practices by deploying advanced analytics and latest, cutting-edge technology, including artificial intelligence. We assist corporations in better understanding their markets and monitoring relevant news.”

The last segment is exhibitions, with RELX using “face-to-face events and digital tools to enabl innovation and supporting the economic development of local markets and national economies around the world.” This segment is supported by the extensive network and reputation of the other branches of the company.

Overall, RELX is not just the individual sub-sections of its business, but a data-first company built over more than a century of focus on accurate, scientific & professional information.

This makes it a good example of the type of business able to thrive in a data-driven age, where AI companies are always looking for more sources of premium-quality data.

For investors, the important story is that this Proprietary Data Moat is now leveraged by the company’s own generative & Agentic AI to provide trusted services in legal, risk prediction, and scientific activities.

Latest RELX PLC (REL) Stock News and Developments

Study Referenced

^{1. Ankur Arora and Rajiv Sabherwal. Do public opinions about ethics, layoffs, and hiring influence stock-market reactions to AI investment announcements? Insights from 2010-2023. The Journal of Strategic Information Systems. Volume 35, Issue 4, December 2026, 101997. https://doi.org/10.1016/j.jsis.2026.101997}