Unity Software Inc.

Current report

Table of Contents

Previous report

AI is becoming an increasingly important part of our strategy and product roadmap, and our failure to successfully develop, deploy, maintain, manage, or commercialize AI-enabled capabilities and products, our reliance on third-party AI models, and the costs associated with such efforts may adversely affect our business, financial results, or competitiveness.

Previous report

AI is becoming an increasingly important part of our broader platform strategy and product roadmap, and our failure to successfully develop, deploy, maintain, manage, or commercialize AI-enabled products could impair our ability to execute that strategy and remain competitive.

Previous report

While we have made, and expect to continue to make, investments to integrate AI into our platform and product offerings, our ability to remain competitive will require increasing levels of such investment over time. There can be no assurance that these investments will enhance our product offerings, improve our competitiveness, efficiency or profitability, or provide an adequate return, and we may not realize the anticipated benefits of these investments in a timely or cost-effective manner, or at all, especially in light of the rapid development of AI technologies in general. In addition, our AI investments and related spending may adversely impact our cost of revenue, research and development expenses, and sales and marketing expenses, which may, in turn, negatively impact our gross margins and operating margins.

Previous report

We generally rely on third-party models for our AI products and features, and our ability to continue to use such technologies at scale depends on access to a limited number of foundational model providers whose availability, pricing, and terms we cannot control. These providers may experience capacity constraints, prioritize other customers, impose usage restrictions, change their terms unfavorably, or terminate their relationship with us, any of which could delay, degrade, or limit the availability of our AI features or make our product offerings less appealing to customers. Our reliance on such models may limit our ability to differentiate our offerings or manage cost and performance characteristics. Customer expectations regarding the speed, availability, and quality of our AI features may also increase over time, requiring us to incur higher infrastructure and model costs (whether for Unity-developed or third party models) that we may be unable to recover through pricing. Any failure to maintain adequate access to third-party AI models on commercially reasonable terms, or to meet rising customer performance expectations with regards to our models and AI features generally, could adversely affect our business, reputation, or financial results.

Previous report

We are increasingly building AI into our offerings, and issues arising from our development or use of AI, or the use of AI by our customers, personnel, vendors, and competitors may adversely affect our business, reputation, or financial results.

Previous report

We are increasingly building AI into our offerings, including Unity Vector, our new machine learning model, our new Unity AI product line, which includes AI Assistant and AI Generators, and Unity Sentis,

Previous report

which allows creators to embed an AI model in the Unity Runtime for their game or application, enhancing gameplay and other functionality, and we are developing new, AI-native products. We continue to advance ML algorithms in our Grow Solutions, which are designed to enable us to provide customers with better performance.

Previous report

AI Assistant and AI Generators leverage third-party models for AI-driven assistance during creation. With respect to third-party models, we do not control the data used to train such models or the specific outputs they generate; we act as a conduit for customer requests and return the responses from those third-party models. The datasets used to train third-party models may inadvertently include content that infringes intellectual property rights. Because we do not control the training data or model weights of third-party models, we may be subject to liability or reputational harm if customers use outputs generated through our offerings that are alleged to infringe third-party intellectual property rights. In addition, customer inputs transmitted to third-party models may include proprietary, personal, or confidential information, and our contractual protections with third-party model providers may not fully mitigate the risk of data exposure, secondary use, or retention of such data by those providers. Proprietary, sensitive, or confidential information of the Company or our customers could be leaked, disclosed, or revealed as a result of or in connection with the use of AI technologies by our personnel or vendors. Unity-developed models have been and will be trained exclusively on directly-licensed or open-licensed data; however, we rely on the accuracy and scope of those licenses as represented by the licensors, and if such representations are incorrect or incomplete, we could face claims that our training data included content used without proper authorization. The outputs of both Unity-developed and third-party models may also inadvertently replicate protected content, resulting in potential claims of infringement.

Previous report

AI technologies are complex and rapidly evolving, and we face significant competition from other companies as well as an evolving regulatory landscape. AI technologies, including generative AI and interactive chatbots, are subject to existing laws of various states and countries, including those regarding privacy, data security, and consumer protection, and our product development efforts may result in new or enhanced governmental or regulatory scrutiny, litigation, ethical concerns, or other complications that could adversely affect our business, reputation, or financial results. Global AI regulatory frameworks remain highly uncertain, rapidly changing, and inconsistent across jurisdictions, and frequently have extraterritorial reach, and, as a result may apply to our AI offerings regardless of where they are developed or deployed. For example, the EU AI Act sets out a risk-based framework, subjecting certain AI technologies to numerous compliance obligations, including transparency, conformity and risk assessment, monitoring and human oversight requirements, and noncompliance could lead to fines of up to 35 million Euros or 7% of our total worldwide annual turnover. Certain provisions of the EU AI Act could require us to alter or restrict our use of AI both in features or products available to our users and in our systems that interact with our users. In addition, certain U.S. states have proposed, enacted, or are considering laws governing the development and use of AI technologies, such as the Colorado Artificial Intelligence Act, California Bot Disclosure Law, the Utah Artificial Intelligence Policy Act, and the California Consumer Privacy Act ("CCPA") regulations on automated decision-making technology.

Previous report

Additionally, certain privacy and data security laws extend rights to consumers and regulate automated decision making in ways that may be incompatible with our use of AI/ML. These obligations may make it harder for us to conduct our business using AI/ML, lead to litigation or regulatory fines or penalties, require us to change our business practices, retrain our AI/ML, delete our models, or prevent or limit our use of AI/ML. For example, the Federal Trade Commission ("FTC") has required other companies to delete algorithms and models derived from or trained on allegedly unlawfully collected data, where it has alleged the company has violated privacy or consumer protection laws. As these requirements continue to emerge or change, we may be required to modify, restrict, or disable certain AI features; implement new technical, governance and other controls; or incur significant compliance and operational costs. Failure to anticipate or comply with evolving AI regulatory obligations could adversely affect our ability to develop and deliver AI-enabled offerings, reduce the competitiveness of our products, or negatively impact our business, reputation, or financial results.

Previous report

Governments are increasingly scrutinizing generative AI technologies for potential intellectual property and content-related harms, and any regulatory enforcement or litigation in this area could

Previous report

adversely affect our business, reputation, or financial results. In addition, AI and ML models may create flawed, incomplete, or inaccurate outputs, some of which may appear correct. This may result from inaccurate, incomplete or flawed inputs, including if a bad actor "poisons" the AI/ML with bad inputs or logic, or if the model’s underlying logic is flawed, resulting in outputs that may appear plausible but are inaccurate (a so-called "hallucination"). We may use AI/ML outputs to make certain decisions. Due to these potential inaccuracies or flaws, the model could lead us to make decisions that could bias certain individuals or classes of individuals and adversely impact their rights. Our AI-enabled technology could be used by an end user in a manner publicly perceived as controversial. The foregoing can cause us to face adverse consequences, including reputational harm, competitive harm, customer loss or legal liability.

Previous report

The increasing use of AI and ML, including AI-driven automation tools or our own AI-based offerings, may enable customers to reduce their reliance on human developers, and therefore paid seat licenses, or otherwise circumvent usage- and seat-based licensing metrics by obfuscating the number of users accessing our products. Any such shift could materially reduce demand for our core subscription-based products and adversely affect our revenues and business model.

Previous report

Advances in AI and 2D and 3D content generation technologies may enable our customers to use AI to automate or replicate aspects of content, content creation, simulation, or runtime functionality provided by our platform and solutions, and allow competitors to develop solutions that reduce the competitiveness of our platform and solutions. Even where such activities are unauthorized or prohibited, our ability to prevent, detect, or stop it may be limited. To the extent such advances reduce the need for our products, or otherwise disintermediate Unity from the content creation process, our business and financial condition could be materially adversely affected.

Previous report

Market acceptance of AI technologies is uncertain, and we may allocate resources toward technologies or products that do not align with customer needs, fail to achieve market acceptance, or are rapidly rendered obsolete by further advances in AI. Our announced or perceived AI roadmap and product strategy may create heightened expectations among customers, and if we fail to meet those expectations, our business, operating results, financial condition, and future prospects could be adversely affected. Our competitors may use AI technologies more efficiently than we do or make technology choices in integrating AI technology or developing AI-based products that prove superior to ours, reducing the competitiveness of our offerings. Decisions as to if and how to integrate or develop AI technologies are difficult, and we, or our customers, may choose, or be required, not to use certain high-performing or lower-cost AI models, including certain open-source models, due to security, compliance, export control, or reputational considerations, which could place us at a competitive disadvantage if our competitors adopt such models without comparable constraints and our business, operating results, and financial condition may be harmed as a result.

Previous report

The planned divestiture of the Supersonic business is dependent on factors beyond our control, including market conditions and identifying a suitable buyer. If a transaction is not completed, is delayed, or occurs on unfavorable terms, we may continue to be exposed to the costs of that business and may need to consider other alternatives.

Previous report

Use of Proceeds

Previous report

None.