Risk topic

Artificial intelligence

730 of 1505 companies added a paragraph in Risk Factors that uses one of these words: AI, artificial intelligence, machine learning, generative, large language model. The paragraph is in the latest 10-K and nothing like it was in the one before. It counts text; it does not say how serious a risk is. Words in capitals only (such as AI) must be in capitals.

By sector

Sector is the SEC's industry description of the company. The bar is the share of that sector's companies compared; each count opens that sector's quotes below.

2247 paragraphs · Fire, Marine & Casualty Insurance show all sectors

ACGL ARCH CAPITAL GROUP LTD. · Fire, Marine & Casualty Insurance

With new technologies and AI tools emerging at a rapid pace, there is no assurance that we will be able to evaluate and integrate new technologies or update our existing systems to keep pace with our competitors and customer needs. While we believe AI presents significant opportunities to support our strategic goals, we may not be successful in implementing AI technologies. It is possible that any AI we use does not perform as anticipated, suffers from “hallucinations” or that its outputs may not be as expected or may result in unlawful discrimination, which may put us at a competitive disadvantage, result in reputational damage and regulatory fines and actions.

Read in the 10-K filed Feb 26, 2026

Additionally, the regulatory landscape surrounding traditional AI and generative AI is evolving, and the expanded use of these technologies may become subject to regulatory scrutiny under new or existing laws. Moreover, the intellectual property and ownership rights associated with both forms of artificial intelligence have not been fully addressed by courts in the U.S. or in other jurisdictions that we operate in. We established the Artificial Intelligence Governance and Oversight Committee (“AIGOC”) to evaluate and approve new AI use cases and issue and oversee our Company’s Artificial Intelligence Policy. While we believe the AIGOC and our larger AI governance framework is responsive to new risks and regulations, failure to comply with the applicable AI-related regulations could result in fines, penalties, litigation, or restrictions on our business operations. These outcomes may have a materially adverse effect on our business or financial condition.

Read in the 10-K filed Feb 26, 2026
AFG AMERICAN FINANCIAL GROUP INC · Fire, Marine & Casualty Insurance

AFG may be exposed to significant risks due to its use or its business partners’ use of AI.

Read in the 10-K filed Feb 25, 2026

AFG uses, and may increasingly rely on, AI technologies in its business operations and may utilize AI in connection with its products and services. AI systems may not perform as intended, may produce flawed, inaccurate, biased, incomplete or otherwise unreliable outputs or analytics, or may be implemented or monitored ineffectively. These issues could cause AFG to write business it would not have otherwise written, misprice policies, assume unintended risks, overpay claims or otherwise experience operational disruptions or financial losses. AFG may incur operational, technological, security, reputational, legal and regulatory risks related to its use of AI. Among other causes, these risks may arise from the misuse or inadvertent disclosure of personal data or sensitive or confidential information; AI-related ethical considerations; vulnerabilities that increase exposure to cyber incidents; failures or limitations in oversight, governance or controls relating to AI systems; or potential intellectual property, contractual or other legal issues associated with AI use. AFG’s business partners, independent agents or other third parties may also develop or utilize AI in their own operations. If such systems fail, malfunction or are improperly designed or monitored, AFG’s operations could be disrupted or there could be direct or indirect adverse impacts to the Company, including to its relationships and reputation.

Read in the 10-K filed Feb 25, 2026

AFG’s existing competitors, new market entrants, technology companies or other third parties may adopt or integrate AI into their business, products and services more rapidly or effectively than AFG, which could make the Company less competitive and negatively impact its results of operations.

Read in the 10-K filed Feb 25, 2026

1 more from AFG, with the rest of what changed in its 10-K

AIG AMERICAN INTERNATIONAL GROUP, INC. · Fire, Marine & Casualty Insurance

•Our development and use of new technology, such as generative artificial intelligence, may present risks.

Read in the 10-K filed Feb 12, 2026

Our development and use of new technology, such as generative artificial intelligence, may present risks.

Read in the 10-K filed Feb 12, 2026

We use artificial intelligence (AI) in our business, including applying generative AI to certain aspects of the underwriting and claims processes in certain lines of business, which may raise technological, security, legal, regulatory and other risks and challenges that may adversely affect our operations, business or reputation. Such risks include the misuse, inadvertent or otherwise, of personal data or other sensitive, confidential or proprietary information, flaws in our or third-party models or training datasets resulting in biased, inaccurate or unanticipated outcomes, ethical considerations regarding the use and deployment of AI technologies, potential infringement of third-party intellectual property rights or the dilution of our intellectual property, and challenges implementing appropriate governance controls to ensure the ongoing, safe deployment of AI systems. The market-wide development of AI tools is growing rapidly and we face competitive risks if our deployment of AI technologies is unable to keep pace with that of our competitors, or we fail to anticipate trends in the use of AI tools.

Read in the 10-K filed Feb 12, 2026

2 more from AIG, with the rest of what changed in its 10-K

KNSL Kinsale Capital Group, Inc. · Fire, Marine & Casualty Insurance

Artificial intelligence is an evolving and rapidly developing technology, and its adoption presents additional risks.

Read in the 10-K filed Feb 20, 2026

opportunities to deploy these tools to aid in decision making processes but cannot guarantee that the risks associated with AI use will be fully mitigated. Any failure to manage these risks could negatively impact our business.

Read in the 10-K filed Feb 20, 2026
L LOEWS CORP · Fire, Marine & Casualty Insurance

Further, CNA’s business could be affected as its policyholders adopt AI technologies. Policyholder use of AI could introduce novel exposures that may result in new or increased claims. Widespread adoption of AI could fundamentally disrupt entire industries, which could impact the demand for certain products.

Read in the 10-K filed Feb 10, 2026

As the cybersecurity threat landscape continues to evolve, Boardwalk Pipelines may be required to expend significant additional resources to continue to modify or enhance its protective measures or to investigate and remediate any information security vulnerabilities. Advances in computer capabilities, discoveries in the field of artificial intelligence, cryptography, inadequate facility security or other developments may result in a compromise or breach of the technology Boardwalk Pipelines uses to safeguard its operational and information technology systems and confidential, personal, or otherwise protected information. As the breadth and complexity of the technologies Boardwalk Pipelines uses continue to grow, including as a result of the use of mobile devices, cloud services, artificial intelligence, open-source software, social media and the increased reliance on devices connected to the internet, the potential risk of cyberattacks and cybersecurity incidents also increases. No security measure is infallible. Despite ongoing efforts to improve its ability to protect its systems from compromise, Boardwalk Pipelines may not be able to protect all of its diverse systems. Boardwalk Pipelines’ efforts to improve security and protect data and its systems may also identify previously undiscovered instances of security breaches or other cyber incidents.

Read in the 10-K filed Feb 10, 2026
LMND Lemonade, Inc. · Fire, Marine & Casualty Insurance

We use artificial intelligence (“AI”) machine learning, and automated decision-making technologies, including proprietary AI and machine learning algorithms and models, (collectively “AI Technologies”) throughout our business, and are making significant investments in this area. For example, we utilize the data gathered from the insurance application process to determine whether or not to write a particular policy and, if so, how to price that particular policy. Similarly, we use proprietary AI Technologies to process many of our claims. The data that we gather through our interactions with our customers is evaluated and curated by proprietary AI Technologies.

Read in the 10-K filed Feb 25, 2026

We expect that increased investment will be required in the future to continuously improve our use of AI Technologies. As with many technological innovations, there are significant risks involved in developing, maintaining and deploying these technologies and there can be no assurance that the usage of or our investments in such technologies will always enhance our products or services or be beneficial to our business, including our efficiency or profitability.

Read in the 10-K filed Feb 25, 2026

In particular, if the models underlying our AI Technologies are incorrectly designed or implemented; trained or reliant on incomplete, inadequate, inaccurate, biased or otherwise poor quality data, or on data to which we do not have sufficient rights or in relation to which we and/or the providers of such data have not implemented sufficient legal compliance measures; used without sufficient oversight and governance to ensure their responsible use; and/or adversely impacted by unforeseen defects, technical challenges, cybersecurity threats or material performance issues, the performance of our business, as well as our reputation could suffer or we could incur liability resulting from the violation of laws or contracts to which we are a party or civil claims. In addition, we may incorrectly price insurance products for our customers or incorrectly pay or deny claims made by our customers, either of which could result in customer dissatisfaction with us, which could cause customers to cancel their insurance policies with us, prevent prospective customers from obtaining new insurance policies, or cause us to underprice policies or overpay claims.

Read in the 10-K filed Feb 25, 2026

9 more from LMND, with the rest of what changed in its 10-K

PLMR Palomar Holdings, Inc. · Fire, Marine & Casualty Insurance

We or our third-party vendors, clients or counterparties may develop or incorporate AI technology in certain business processes, services or products. Our research and development of such technology remains ongoing, and AI algorithms and training methodologies may be flawed. Leveraging AI capabilities to potentially improve our internal operations also presents further risks, costs, and challenges. While we aim to develop and use AI responsibly and attempt to identify and mitigate ethical and legal issues presented by its use, we may be unsuccessful in identifying or resolving issues before they arise.

Read in the 10-K filed Feb 24, 2026
ROOT Root, Inc. · Fire, Marine & Casualty Insurance

•we fail to provide effective updates to our existing products or to keep pace with technological improvements in our industry, including generative AI and machine learning technology;

Read in the 10-K filed Feb 25, 2026
SKWD Skyward Specialty Insurance Group, Inc. · Fire, Marine & Casualty Insurance

Artificial intelligence is an evolving and rapidly growing technology which may impact our business and operations.

Read in the 10-K filed Mar 2, 2026

The rapid growth and development of artificial intelligence and machine learning may alter the competitive landscape in which we operate. Our employees utilize artificial intelligence for risk selection, pricing and claims handling to aid in their effectiveness and efficiency and we continue to research and implement artificial intelligence-based solutions in an effort to improve our business. Notwithstanding this, our competitive position may be harmed if competitors are able to leverage artificial intelligence solutions more quickly or more effectively. In addition, while we do not rely solely on information and analysis production through artificial intelligence in our decision making, if the content, analyses or recommendations that artificial intelligence applications assist in producing are, or are alleged to be, deficient, inaccurate or biased, such as due to limitations in algorithms, insufficient or biased base data or flawed training methodologies, our business, financial condition, results of operations and reputation may be adversely affected. Further, as artificial intelligence technology and products are continuously evolving, we may incur costs to adopt and deploy technologies that could become obsolete earlier than expected. There can be no assurance that we will realize the desired or anticipated benefits from artificial intelligence.

Read in the 10-K filed Mar 2, 2026

There is also uncertainty in the legal and regulatory landscape at the federal and state levels for the use of artificial intelligence, which is not fully developed. Any laws, regulations or industry standards adopted in response to the emergence of artificial intelligence may be burdensome, could entail significant costs, and may restrict or impede our ability to successfully develop, adopt and deploy artificial intelligence technologies efficiently and effectively.

Read in the 10-K filed Mar 2, 2026
SPNT SiriusPoint Ltd · Fire, Marine & Casualty Insurance

A significant amount of communication between our employees and our business banking and investment partners depends on information technology and electronic data exchange. We also license certain systems, platforms, datasets and, increasingly AI-driven tools and models from third parties and rely on cloud-based infrastructure and externally hosted systems. We cannot be certain that we will maintain continued access to these systems, platforms, datasets or models, or that they will operate as intended. AI models may produce inaccurate, biased, or unpredictable outputs, may degrade over time, or may require retraining based on data we do not control. We also cannot guarantee that we could replace these tools without degrading operational performance or slowing our underwriting or claims response time. In addition, if we adopt an overly cautious approach in evaluating, updating, or integrating new technology or AI‑enabled systems, we may delay necessary improvements, prolong exposure to outdated or less secure systems, or impede operational efficiency, which could increase our vulnerability to cybersecurity incidents or other technology‑related failures.

Read in the 10-K filed Feb 24, 2026

Like all companies, our information technology systems are susceptible to interruptions or failures related to events beyond our control, including natural disasters, terrorist attacks, third party service outages, and general technology failures. As AI technologies become more integrated into our operations, failures, errors or unexpected behavior in these systems could also

Read in the 10-K filed Feb 24, 2026

We are subject to evolving cybersecurity, data protection, and AI-related laws and regulations across the jurisdictions in which we operate. Regulatory expectations for cybersecurity governance, AI governance, reporting, and operational resilience continue to increase globally. Compliance with new or developing requirements — including AI model documentation, data governance controls, risk assessments, testing, third‑party oversight, breach notification, and incident response protocols — may increase our compliance costs and operational burdens. As our operations expand into additional jurisdictions and as the regulatory landscape for AI and automated decision-making evolves, we expect that our cybersecurity, data privacy, and AI compliance costs will continue to rise.

Read in the 10-K filed Feb 24, 2026
THG HANOVER INSURANCE GROUP, INC. · Fire, Marine & Casualty Insurance

Our inability to adapt to or implement new technologies, including artificial intelligence, could adversely affect our business and results of operations, and the use of new technologies may create unforeseen exposure or coverage issues.

Read in the 10-K filed Feb 20, 2026

We rely on a broad range of advanced technologies, including AI, to support our operations. As these technologies continue to rapidly evolve, our business could be adversely affected if we are unable to effectively adopt, update, integrate, or scale new capabilities in a timely or cost-effective manner. Competitors with greater technology investments or faster innovation cycles may introduce enhanced AI enabled tools, digital experiences, or analytic capabilities that improve efficiency, the customer or agent experience, and business outcomes. If we fail to keep pace with such developments, we may face disadvantages in attracting and retaining policyholders, supporting our distribution partners, or achieving operational efficiency, which could adversely affect our business and results of operations. New or evolving technologies, including AI, may also create unforeseen exposures or coverage issues under the policies we write or introduce new forms of claims fraud or cybercrime. In addition, regulatory standards relating to the use of AI in the states where we do business are also evolving and may increase the complexity of our compliance and reporting obligations, as well as the regulatory risks associated with our use of this technology.

Read in the 10-K filed Feb 20, 2026

insider threats, software vulnerabilities, and operational or system failures. A cybersecurity attack or other security incident affecting us or a third-party service provider could disrupt our ability to quote, bind, issue, or service policies, impede claims intake, investigation, or settlement, compromise or misappropriate confidential agent, policyholder, claimant, or commercial customer information, affect our catastrophe response capabilities, or interfere with the operation of telematics, Internet of Things devices, or other data collection tools used in underwriting or claims processing. Such events could also expose us to regulatory investigations or enforcement actions by state insurance departments, litigation, contractual liabilities to distribution partners, remediation costs, operational delays, reputational harm, and adverse impacts on our financial condition or results of operations. Although we maintain a cybersecurity program designed to prevent, detect, and respond to security threats and continually enhance our defenses, no control environment can eliminate all risk. Like many companies in the property and casualty insurance industry, we and our third-party service providers have experienced, and are likely to continue to experience, cybersecurity events and attempts to compromise systems. To date, these events have not had a material adverse effect on our business, but there is no assurance that future incidents will not be material. Cybersecurity threats continue to grow in frequency and complexity, including due to the rapid advancement and use of AI by threat actors. As a result, we may be required to devote additional financial, technological, and personnel resources to monitor, detect, investigate, contain, mitigate, remediate, or defend against such threats. These efforts may increase costs and divert resources from other operational or strategic priorities.

Read in the 10-K filed Feb 20, 2026
Esc

Type a ticker such as AAPL, a name, an investor such as Buffett, or a topic such as tariffs. ↑ ↓ to move, Enter to open.