Monthly Archives: March 2025

The Innovation Dilemma: AI Distillation in OpenAI v. DeepSeek

The legal battle between OpenAI and DeepSeek has ignited heated debate about artificial intelligence innovation, intellectual property rights, and competitive dynamics in the AI industry. OpenAI–the undisputed industry leader in generative AI backed by billions in funding–alleges that DeepSeek violated its terms of service by leveraging a technique known as “distillation” to build a competitive product.

Distillation is a method in AI development that enables a smaller “student” model to replicate or approximate the performance of a larger “teacher” model by learning from its outputs. Due to the method’s ability to lower computational costs, distillation has become a widely regarded technique for creating AI systems more efficiently. However, OpenAI claims that DeepSeek’s approach—reportedly querying OpenAI’s model at scale and using its responses as training data to improve DeepSeek’s own AI—crossed a line by extracting OpenAI’s proprietary data without permission.

If OpenAI’s allegations lead to a legal battle, the outcome could have a substantial impact on the future of the generative AI industry. This case highlights a recurring dilemma in the tech industry: how to balance the right innovators have to protect their technological advancements against the broader public interest in ensuring open access to transformative technologies. A notable example of this was the legal dispute between Oracle and Google over the use of Java APIs in the Android operating system. Oracle sued Google, alleging that functional software elements (such as APIs) are subject to copyright protection; thus, the replication of Java APIs on Android constituted copyright infringement. In 2021, the U.S. Supreme Court ruled in favor of Google, determining that Google’s use of the Java APIs was a lawful fair use. The Court emphasized that certain forms of software replication can drive innovation rather than hinder it, outweighing Oracle’s desire to control its copyrighted software.Had Oracle prevailed, developers might have faced significant restrictions on API usage, potentially stifling interoperability and innovation within software ecosystems.

A similar dynamic is playing out in the AI industry. OpenAI’s massive financial and research investments, including a recent $6.6 billion funding round, have driven the development of advanced models like GPT-4.5. A ruling in favor of OpenAI would strengthen the legal protection that companies with AI models have, preventing other companies from using distillation. This would provide firms with greater legal assurance that their technological advances cannot be easily reproduced, encouraging further investment in AI research. Companies with large-scale computational infrastructure, which would benefit from a more predictable and enforceable intellectual property landscape, would likely be incentivized to expand their AI initiatives.

Additionally, such a ruling could pave the way for industry-wide licensing frameworks, where AI firms must obtain explicit permissions or pay for access to use large proprietary models rather than extracting their outputs through distillation. This shift could lead to a more standardized business model in AI, allowing companies to commercialize access to their models while ensuring that smaller firms can still legally participate through licensing agreements. As a result, AI developers could receive financial compensation for their innovations, creating a sustainable ecosystem where AI research is funded without fear of imitation.

Conversely, if distillation is considered a legitimate practice, it would make the development of advanced AI accessible to a wider range of companies. This would drastically change the dynamics of the AI industry. Currently, building state-of-the-art AI models requires enormous computational resources, access to vast datasets, and significant financial backing. These barriers make it nearly impossible for smaller startups or independent researchers to compete with tech giants like OpenAI, Google DeepMind, and Anthropic, which have billions of dollars in funding and access to specialized hardware such as high-end GPUs and TPUs.

By allowing distillation as a legal practice, smaller AI companies could train efficient models using knowledge extracted from larger, more advanced AI systems without having to replicate the expensive training process from scratch. Instead of needing to collect and process massive datasets—often a key advantage held by large companies—smaller firms could leverage the distilled knowledge from publicly available models or even commercial APIs to develop lighter, more cost-effective AI systems tailored to specific use cases.

For example, a healthcare AI startup that lacks the resources to train a large-scale medical language model from the ground up could apply distillation techniques to a commercially available model to develop a specialized AI assistant for doctors. Similarly, a legal tech firm might use distillation to fine-tune an AI system focused exclusively on contract analysis, making legal AI tools more affordable and widely available to law firms and in-house legal teams.

Distillation could thus encourage AI adoption in industries that large tech companies typically overlook. While OpenAI and Google DeepMind focus on broad, general-purpose AI systems, smaller companies could use distilled models to create highly specialized AI solutions for niche markets such as agriculture, local governance, small business automation, and environmental monitoring. These applications might not be financially viable for large tech firms to pursue but could thrive in a more open AI ecosystem where smaller players have access to efficient AI development techniques.

This increased accessibility would create a more competitive AI landscape, as more companies could afford to develop their own models without relying on a handful of dominant firms for licensing access. Instead of AI advancements being controlled by a few large corporations, startups and independent developers would have more opportunities to innovate, leading to faster technological progress and more diverse AI applications across different industries.

Ultimately, if distillation remains a widely accepted practice, it would democratize AI innovation, making it more feasible for smaller players to enter the industry and compete with established giants. This shift would not only increase competition and drive down costs for AI-powered products and services, it would also lead to a more diverse and inclusive AI landscape. Innovation would not solely be driven by a few massive corporations but rather by a global network of researchers, startups, and independent developers working on AI applications that address a wide range of real-world challenges.

The ongoing debate surrounding the potential legal dispute between OpenAI and DeepSeek encapsulates a pivotal moment in the evolution of artificial intelligence. This situation compels stakeholders to weigh the benefits of democratizing technology against the necessity of protecting significant investments and maintaining industry standards. As discussions progress, the implications are likely to ripple through the AI industry, potentially reshaping the landscape of innovation, market dynamics, and regulatory policies for years to come.

Are Board Diversity Rules Coming to an End?

In recent years, corporate boardroom diversity efforts have gained momentum. This shift is fueled by growing recognition of the benefits that diverse perspectives bring to decision-making and corporate performance. However, there have been significant setbacks for the diversity of corporate boards in the US in recent months, strengthened by President Trump’s Executive Order terminating government DEI programs. For instance, Goldman Sachs’ decision to reverse its initial public offering (IPO) diversity mandate and the Fifth Circuit Court of Appeals’ decision to invalidate Nasdaq’s Board Diversity Rules both point to a potential change in corporate governance. Although these occurrences underscore the growing legal and commercial challenges to mandatory diversity programs, one key question remains: Will firms continue their efforts through voluntary means and market-driven incentives, or is the era of board diversity regulations really coming to an end?

The Fifth Circuit’s Decision on Nasdaq’s Board Diversity Rules

Originally approved by the Securities and Exchange Commission (SEC) in 2021, the Nasdaq Board Diversity Rules required that listed businesses report the diversity makeup of their boards and include at least one female member and one member from an underrepresented minority or the LGBTQ+ community. Companies that failed to meet these criteria were mandated to explain their reasoning.

The Fifth Circuit Court of Appeals, however, recently declared these regulations as unconstitutional. Although Nasdaq operates as a self-regulatory organization, the court ruled that it exceeded its authority by imposing rules that effectively demanded businesses to comply with requirements similar to affirmative action.  The court also questioned the SEC’s approval of the rule, noting that the agency failed to demonstrate how the standards aligned with the Securities Exchange Act’s objectives, such as protecting investors, preventing market manipulation, and promoting competition. As a result, the Fifth Circuit ruled that the SEC exceeded its regulatory authority under the Act.

The court concluded that requiring disclosure of gender and race diversity on boards did not directly safeguard investors, and the comply-or-explain mandate had little to do with the main objectives of the Act. Through this action, the court applied the “major questions doctrine,” which holds that federal agencies must have clear congressional authorization for significant regulatory actions. Therefore, it ruled that the SEC lacked the explicit authority from Congress to regulate corporate board structures, particularly in politically sensitive areas. In other words, the court reasoned that the SEC had overreached its customary regulatory purview, which is centered on market manipulation and proxy voting, by approving Nasdaq’s Diversity Rules, infringing on issues that are under the purview of other authorities.

Goldman Sachs’ Reversal of IPO Diversity Requirements

By declaring in 2020 that it would not underwrite IPOs for businesses without at least one diverse board member, Goldman Sachs established itself as a pioneer in corporate diversity initiatives. The company’s dedication to enhancing corporate governance was demonstrated in 2021 when it increased the need to at least two diverse directors.

However, Goldman Sachs recently retracted this requirement in response to growing political and legal scrutiny. The firm cited client feedback and changing market conditions as justifications for dropping the mandate. The reversal implies that financial institutions are being more cautious about diversity standards out of concern about potential legal challenges or investor reaction against required measures.

The Road Ahead for Board Diversity

Institutional investors and shareholder activism will play a crucial role in shaping the future of board diversity. Beyond financial institutions withdrawing their diversity mandates, major institutional investors are now revising their board diversity policies. BlackRock’s 2025 policy removes numerical diversity targets, eliminating its prior disclosure-based proxy vote guidelines. However, it retains the option to take voting action against S&P 500 boards that deviate significantly from market norms. Vanguard has similarly softened its stance, replacing explicit gender, race, and ethnicity requirements with a broader emphasis on “cognitive diversity” and skill. However, it maintains the right to vote against nominating committee chairs if diversity-related disclosures are inadequate. State Street has taken the most significant step back by entirely removing numerical diversity targets and no longer voting against boards that fail to meet prior diversity standards. Instead, it now emphasizes that nominating committees should oversee board composition. In comparison, the global company BNP Paribas is strengthening its diversity policies for 2025, raising its gender diversity target to 40% female directors. The Fifth Circuit’s ruling, Goldman Sachs’ revised policy, and shifting institutional investor behaviors raise major questions about the future of board diversity. Although regulatory mandates for board diversity might be diminishing, the need for diverse leadership is expected to persist. The Fifth Circuit’s decision could create a precedent that dissuades exchanges or regulators from enforcing diversity mandates, causing companies to be reluctant in establishing formal diversity goals because of legal concerns.

However, other firms might see these changes as a chance to adopt a more proactive and voluntary strategy. Companies can advance their diversity objectives while staying within legal and regulatory boundaries by fostering a culture of belonging, setting internal diversity targets, and implementing inclusive hiring practices. Moreover, companies can gain an advantage in attracting elite talent, fostering customer loyalty, and enhancing their brand by committing genuinely to diversity and inclusion.

These challenges for board diversity mandates do not necessarily indicate the end of corporate diversity initiatives. Market-driven pressures, institutional investor activity, and voluntary corporate initiatives can sustain the momentum for diversity and inclusion, even though legal and regulatory restraints may limit the use of quotas and requirements. The attitude of businesses, investors, and stakeholders to value diversity as a moral and business necessity will ultimately determine the future of board diversity.

As the corporate world navigates this evolving challenge, one thing is clear: the pursuit of diversity and inclusion is far from over. The objective of making boardrooms more inclusive and equitable is still as crucial as ever, whether it is achieved through volunteer efforts or legal requirements. In light of tackling these challenges, the question is not whether board diversity is ending but rather how businesses can innovate and adapt to sustain diversity.

Back to the Boardroom: SEC Restricts Shareholder Proposals

Under Rule 14a-8 of the Securities Exchange Act of 1934, shareholders are entitled to submit proposals for company action for inclusion in a company’s proxy statement to be voted on at annual shareholder meetings. In theory, shareholder proposals empower investors to shape corporate action by putting plans shareholders wish to see the company implement to a vote. While not always binding, if the board approves a proposal and it has enough shareholder support, shareholders can put significant pressure on the board to follow the proposal’s advice. Recent years saw a trend in many proposals focused on environmental, social, and governance (ESG) concerns, often conflicting with corporate management’s preferences.

On February 12, 2025, the Division of Corporation Finance of the U.S. Securities and Exchange Commission (SEC) rescinded Staff Legal Bulletin No. 14L (SLB 14L) and issued Staff Legal Bulletin No. 14M (SLB 14M), making it easier for public companies to exclude ESG shareholder proposals from being voted on at annual meetings. SLB 14M rolls back the policy introduced by SLB 14L in 2021, which made it “markedly easier for shareholders to put certain environmental and social proposals to vote” by adopting an expansive view of shareholders’ ability to raise concerns about a company’s social policies. This decision reverts SEC policy to the stance it held during President Trump’s first term, tracking a notable shift in how companies and investors engage in corporate governance: redistributing discretion from shareholders back to the boardroom.

Rule 14a-8 and the SEC No-Action Letter

Companies rely on Rule 14a-8(i)—particularly, Rule 14a-8(i)(5) (the “economic relevance” exclusion) and Rule 14a-8(i)(7) (the “ordinary business” exclusion)—to exclude a proposal from a shareholder vote. If a company wishes to exclude a proposal, it may request SEC staff to comment on whether the rules permit the omission through a “no-action letter.” If granted, the company can exclude the proposal without fear of enforcement action.

SLB 14L’s broad interpretation of these exclusions enabled shareholders to put to vote proposals that normally would be subject to exclusion. SLB 14M re-narrows the scope and application of these exclusions, thus limiting the power of shareholder proposals.

Rule 14a-8(i)(5): The “Economic Relevance” Exclusion

Rule 14a-8(i)(5) allows for the exclusion of proposals that (1) relate to operations which make up less than 5% of the company’s total assets, profit, and gross sales at the end of the last fiscal year and (2) are not otherwise significantly related to the company’s core business. Under SBL 14L’s broad interpretation, proposals that did not meet the economic threshold requirements (and thus did not comply with the rule’s economic concern standard) but raised “broad social or ethical concerns” could not be excluded. Staff would not issue no-action letters for such proposals—regardless of how small the operations were—thus prioritizing corporate responsibility above concerns of shareholder micromanagement. SLB 14M disregards the importance of social and ethical issues “in the abstract,” instead analyzing the economic relevance of the proposal to the company’s business on a case-by-case basis. This would make it effectively impossible for shareholders to address ESG concerns in their proposals unless those concerns significantly affect key aspects of the company’s core business.

Rule 14a-8(i)(7): The “Ordinary Business” Exclusion

Under Rule 14a-8(i)(7), a company may exclude proposals that deal with the company’s “ordinary business operations” on the basis that certain day-to-day tasks should be decided by management rather than being subject to direct shareholder oversight. SLB 14L’s expansive interpretation of “significant social policy” meant that companies couldn’t exclude shareholder proposals that related to day-to-day business matters so long as the proposal also raised broad social concerns—regardless of whether there was a nexus between the policy issue and the company. As a result, “during the 2022 proxy season after SLB 14L’s issuance, proposals concerning environmental topics increased over 50%.” SLB 14M reverts to a pre-SLB 14L “company-specific approach,” with the staff now allowing exclusions when a policy issue is not significant to a specific company. This limits shareholders’ ability to raise ESG proposals that bear on managerial discretion regarding business operations.

Limiting Shareholder Micromanagement

Lastly, SLB 14M also reinstates restrictions on shareholder micromanagement by allowing companies to exclude proposals that (1) seek intricate detail or specific timeframes for implementing complex policies and (2) are highly prescriptive, supplanting management discretion in decision-making. For instance, the bulletin explicitly states that proposals to reach net-zero emissions by a given year are now excludable as micromanagement because they impose timelines and limit management’s flexibility.

Effects on Industry

The Commission’s move follows a sharp decline in investor support for ESG-related shareholder proposals, having dropped from 21% in 2021 (the year SLB 14L was issued) to just 3% by 2023. In 2024, only four (1.4%) out of 279 assessed proposals received majority support. The trend marks a growing resistance to ESG-driven shareholder activism. Considering that six of the largest banks in the U.S.—Citigroup, Bank of America, Morgan Stanley, Wells Fargo and Goldman Sachs—already withdrew from the UN-sponsored net-zero emissions banking alliance weeks before the SEC’s announcement, the new bulletin likely marks the proverbial nail in the coffin for an era of investor-driven environmental impact.

The release of SLB 14M just as the current proxy season gains momentum makes it even harder to hold the world’s largest companies accountable. Many shareholder proposals this year focus on ESG issues that investors believe have a material impact on a company’s financial value. However, a significant number of these proposals will likely be excluded from proxy ballots because SLB 14M was issued too late for investors to adjust their proposals to fit its stricter standards. SEC Commissioner Caroline A. Crenshaw is concerned that the timing of this guidance disproportionately benefits corporations, allowing them to revise their no-action requests to exclude proposals while leaving investors without a similar opportunity to adapt.

The full effect of these changes is unlikely to be seen until next proxy season, but for now, SLB 14M marks a shift toward greater corporate control over shareholder proposals. This will likely lead to a decline in ESG-related initiatives and perhaps a shift toward alternative shareholder advocacy strategies like direct engagement or even litigation. Future regulatory or political changes could reshape these rules again, but as it stands, shareholder activism faces a more restrictive landscape.

The Last Frontier: Why Private Equity Wants a Piece of Law Firms

What do trailer parks, hospitals, prison services, and accounting firms have in common? Private equity (PE) has acquired a stake in all of them—and it’s not stopping there. From 2023 to 2024, PE firms closed $565 billion in announced deals, a 25% increase in value. With so much capital in play, PE is ever on the lookout for its next conquest—and now it has set its sights on a domain long considered off-limits: law firms. Historically, regulatory and ethical barriers have kept non-lawyer investors out of the legal industry. Yet if there’s one thing we know about PE, it’s that where big money stands behind a closed door, they will find a way to pry it open.

PE potentially views law firms as a way to “print money.” As David Marcus writes in The Allure of Law Firms for Private Equity, “Recent investments by [PE] in accounting firms suggest PE would find attractive targets in the legal profession if more state bar associations allow nonlawyers to own stakes in law firms.” Indeed, over the past few years, PE sponsors have acquired major stakes in several top accounting firms, including one deal that reached a valuation of $2 billion. Law firms, like these accounting firms, have “reliable cash flows and low capital expenditures​​,” making them prime candidates for PE. However, one thing stands in PE’s way—the American Bar Association (ABA).

The ABA, whose legal standards most states follow, created Rule 5.4—formally titled “Professional Independence of a Lawyer”—to ensure that client interests, not outside investors, dictate legal decisions. By prohibiting fee-sharing and partnerships with non-lawyers, the rule aims to maintain broad ethical standards and foster public confidence that legal advice remains free from business imperatives. It also protects client confidentiality by preventing outside stakeholders from accessing sensitive information. The core idea is that without external investors, lawyers are bound solely to provide high-quality, conflict-free representation—not to maximize returns for shareholders.

Yet critics of Rule 5.4 argue it’s a double-edged sword: While it guards against undue profit-driven influence, it also prevents law firms from tapping outside capital. Scholars at Stanford Law point out that the United States, despite hosting one of the largest pools of lawyers in the world, still ranks abysmally—109th out of 128 countries—in access to affordable civil legal services. They see Rule 5.4 as a prime suspect, explaining that the lack of external funding leads to limited investment in various areas, such as new technology solutions, streamlined business structures, and wider consumer outreach. Tom Lenfestey of the Law Practice Exchange agrees that sponsor-backed funding could modernize firms and improve services for underrepresented clients. Meanwhile, PE firms see themselves as the ones capable of bringing these efficiencies to the legal field, but only if Rule 5.4 is reformed.

Advocates for reform can now point to the last few years for evidence that Rule 5.4 is unnecessary, with states like Arizona and Utah successfully bypassing it. In 2020, Utah launched a regulatory sandbox, allowing non-lawyers to hold ownership stakes under close judicial oversight. The program has proven successful enough that the Utah Supreme Court extended it through 2027. Arizona went even further by eliminating its ban on non-lawyer ownership of law firms. Under its Alternative Business Structure (ABS) framework, approved entities must meet stringent requirements such as licensing, malpractice insurance, and background checks on owners to ensure professional standards remain intact. So far, over 100 ABS licenses have been granted, reflecting Arizona’s enthusiasm for this new model. Proponents say such reforms could help expand access to legal services and lead to lower legal fees through expanded legal access. For now, though, Rule 5.4 still holds in most U.S. jurisdictions, keeping PE’s capital, expertise, and ambitions largely at arm’s length.

At first glance, loosening Rule 5.4 seems promising: Let outside investors pour in capital, achieve economies of scale, and serve the communities we’ve historically neglected. But if PE’s track record in healthcare and accounting is any guide, there’s also an alarming side. A 2023 study has even shown that Medicare patients at PE-owned hospitals suffered a 25% increase in complications, and many accountants lament losing their independence. Why would lawyers be immune? If non-lawyer owners offer large buyouts to senior partners, firms could soon be hardwired to prioritize maximizing returns over their duty of loyalty to clients.

Yet a total meltdown isn’t guaranteed. Arizona and Utah have tested regulatory sandboxes and Alternative Business Structures, and so far, there has been no wave of ethical collapses. Moreover, not every PE fund is incentivized to focus only on maximizing profits, like in dentistry, where PE firms have preferred generating steady returns and cultivating a mutually beneficial relationship where they help manage the nonclinical aspects of the practice.

That’s the tension: Rule 5.4 has shielded client interests from outside meddling for decades. However, critics argue that it stifles innovation, pushing America’s legal costs so high that it’s 109th worldwide in affordability. So, do we commence reforms despite the risk of replaying a corporate exploitation script?

It all depends on the details of any reform. Tear down Rule 5.4 wholesale, and sure, we could see cynical cash grabs, gutted ethics, and vulnerable clients left behind. But it’s naive to pretend existing ethics rules can fix the dearth of legal access, especially when in 2023 alone, of the 55% of Californians who experienced a civil legal problem, only 30% received legal assistance. If Arizona and Utah’s experiments hold up—if real guardrails keep investors from hollowing out the soul of legal practice—maybe we can harness new capital and provide the American public with better access to legal services.

In short, Rule 5.4 isn’t just an “obstacle,” nor is its reform a “silver bullet.” It’s a decades-old ethical foundation that deserves scrutiny and shouldn’t be dismantled without care. Whether the drumbeat of expanding legal access is just a Trojan horse for corporate profit or a legitimate reform that could lift more boats ultimately depends on us: how we write the regulations, how we hold investors accountable, and how we preserve the independence and integrity of law as a profession.

AI in Biotechnology: Antitrust Issues with Big Data, Data Monopolization, and Exclusive AI Licensing

Artificial intelligence (“AI”) is transforming every aspect of our lives, and the biotechnology field is no exception. It drives significant advancements across various areas, such as drug discovery and development, genetic research, personalized medicine, and medical diagnostics. These innovations rely on vast amounts of large complex datasets, often called big data, both as input for AI models and as a key component of the research and development process. However, as AI continues to rise, some market players are concerned about the implications of data control and the potential impact of AI on the future of the field.

Major concerns surround big data and data monopolization from an antitrust and competition perspective, even outside the biotechnology field. While big data drives innovation and enhances consumer offerings, the Federal Trade Commission (“FTC”) has highlighted that big data can be used to discriminate against and target minorities, potentially excluding them from certain services and products. Additionally, restricted access to valuable data, particularly when it is costly or difficult to generate, can create entry barriers that limit competition and innovation. This phenomenon is often termed data monopolization: dominant players in a market controlling vast amounts of data, determining its use, and, in doing so, sustaining their market dominance.

However, determining whether data control is an entry barrier and thus anti-competitive is complex, as much of the information contained in these datasets is not unique; it can be obtained through alternative means. For example, the FTC and state attorneys general sued Facebook in 2020 for alleged monopolization. Facebook was accused of limiting data access to non-competing developers, deterring the rise of rival platforms. However, the court dismissed these claims under Trinko’s no-duty-to-deal rule, which holds that antitrust laws do not require a firm to engage with its competitors. The court reasoned that forcing Facebook to share its data could risk its incentives for innovation, place the judiciary in the role of a regulator, and potentially lead to collusion with its rivals.

This precedent may shape how antitrust authorities will assess the implications of big data and data monopolization, eventually limit data sharing and impact regulatory efforts. From an AI perspective, it could also hinder smaller companies from training their models and entering the market.

Furthermore, in the biotechnology sector, data monopolization might become an even more hotly-contested issue than it currently is in the larger tech industry. Many companies conduct trials and collect personal health data, which can be challenging to generate and typically requires obtaining individual consent before any use. These challenges are particularly salient for companies with AI tools that collect and process large volumes of specialized data, such as genetic information, which is crucial for developing new medicines and medical products. Smaller companies often lack the resources to access such unique data, potentially halting their research and preventing them from entering the market. In this context, biotech startups may argue that the uniqueness of this data creates a significant barrier to competition. However, courts and regulators in the future may have to consider the types of health data collected, which data should be eligible for sharing, and whether large companies should be required to share years of experiments and proprietary data that may constitute intellectual property.

Another antitrust concern is AI-exclusive licensing. Generally, FTC permits exclusive licenses, a standard market practice where an intellectual property owner grants exclusive usage rights to a licensee. However, concerns arise when a dominant manufacturer leverages such agreements to make it more difficult for smaller competitors to compete effectively. Exclusive contracts can also limit access to lower-cost suppliers, compelling competitors to source from more expensive alternatives, which may influence market conditions.

In the context of AI, exclusive licensing can raise several issues regarding market competition. First, as AI relies heavily on data, companies that control essential datasets may choose which entities to contract with and license the data to, potentially limiting access for competitors. Second, a dominant company may grant exclusive rights to use its AI technology to another major player, also excluding smaller competitors by forming selective partnerships. Third, an established company might suppress competition by bundling its AI products with proprietary software, favoring itself and preferred partners while disadvantaging new entrants.

For example, the European Commission is investigating whether Google’s exclusivity agreements with device manufacturers, such as Samsung, to pre-install its AI model could limit competition by giving its AI model a default advantage, as users are more likely to stick with the pre-installed option rather than explore alternatives. While restricting such exclusivity agreements could potentially hinder innovation by limiting direct channels for major players’ products, it could also lead to an AI market dominated by a few giants, potentially sidelining smaller competitors.

Again, these concerns are particularly exacerbated given the highly competitive nature of the biotechnology sector, which only a few major pharmaceutical companies dominate. Large pharmaceutical firms, even before the advent of AI, have often collaborated with other industry leaders to bundle their products together, leveraging their extensive resources and diverse offerings. With the growing importance of data and AI, these firms may secure exclusive data licenses and access to critical data from hospitals and other health providers, which could disadvantage smaller startups that lack the same level of access. AI models in the biotechnology sector could be crucial for identifying drug candidates, personalizing treatments, and predicting clinical outcomes. Exclusive AI licensing could restrict access to these advancements, limiting competition and creating significant barriers for emerging biotechnology startups.

As AI continues to influence the biotechnology sector,  its implications could be far-reaching. big data and AI raise legitimate antitrust concerns, particularly in an environment where major pharmaceutical companies have significant advantages over smaller startups in terms of resources and products. These larger companies could leverage their position to maintain market power. While innovation continues to progress, antitrust authorities must closely monitor the industry and intervene when barriers to competition arise, given the sector’s crucial impact on public health and well-being. As these concerns increase, the U.S. enacted the 21st Century Cures Act to limit information blocking in healthcare. Though it does not mandate proactive data sharing, its role may grow as antitrust enforcement evolves.