More than half of companies investing in artificial intelligence have not yet established governance policies to manage the technology responsibly, according to a new S&P Global report. As AI adoption accelerates across industries, the focus is shifting from whether companies can deploy sophisticated AI systems to whether they can do so in a way that protects data, minimizes bias, and addresses environmental concerns.

Why Are Companies Falling Behind on AI Governance?

S&P Global's third-quarter 2026 Sustainability Quarterly found that while firms have invested heavily in AI-enabled tools, most lack the guardrails to manage them responsibly. The Corporate Sustainability Assessment (CSA) revealed that 54% of firms have yet to put AI governance into practice, even though the share of companies with some form of AI policy has grown from 40% in 2024 to 47% in 2025.

Among companies that do have policies in place, the focus areas tell an important story about where governance efforts are concentrated and where critical gaps remain:

• Data Privacy: 91% of firms with AI policies have focused on protecting personal data and ensuring compliance with privacy regulations.
• Bias Avoidance: 74% have implemented policies to identify and reduce algorithmic bias in AI systems.
• Cybersecurity: 69% have established safeguards against security threats and unauthorized access.
• AI-Generated Content Detection: Only 40% have policies in place to identify content created by AI, which S&P Global describes as a "clear governance gap."

The governance lag is particularly concerning given the rapid expansion of AI across sectors. About 41% of companies in the analysis do not have a policy at all, while another 13% said they do not currently have a policy but plan to implement one within the next two years.

How Is AI's Energy Consumption Affecting Climate Goals?

One of the most pressing challenges highlighted in the report is the energy demands of AI infrastructure. Among firms operating or leasing data centers, power consumption nearly tripled between 2017 and 2024, reflecting the explosive growth in AI workloads. However, there is a bright spot: renewable energy use has grown in tandem, accounting for 84% of total energy consumption by 2024.

Despite these gains, the report found a troubling disconnect. While many companies are using AI tools to improve sustainability, most are not measuring whether those efforts actually work. According to S&P Global's CSA, just 30% of firms have quantified the impact AI is having on addressing environmental concerns.

Energy efficiency is the most common environmental application of AI across sectors, with about 27% of companies saying they are using AI to measure or improve climate-related metrics such as emissions. Yet without rigorous measurement, the true climate benefits of these AI-driven programs remain largely anecdotal.

Steps to Strengthen AI Governance in Your Organization

• Establish a Dedicated AI Policy: Create a standalone AI governance framework or integrate AI oversight into existing governance documentation. This should address data privacy, bias detection, cybersecurity, and content authenticity verification.
• Measure AI's Environmental Impact: Implement systems to track and quantify the climate benefits of AI-driven sustainability programs. Set clear metrics for energy efficiency, emissions reduction, and renewable energy use in data center operations.
• Develop AI-Generated Content Standards: Close the governance gap by establishing policies to identify, label, and manage AI-generated content across your organization, particularly in customer-facing and regulatory contexts.
• Monitor Data Center Energy Use: If your organization operates or leases data centers, track power consumption trends and set renewable energy targets aligned with your climate commitments.

What Does Community Pushback Mean for AI Expansion?

Beyond governance and energy concerns, the report highlights growing public resistance to AI infrastructure. Around 75% of Americans now oppose the construction of data centers near their homes, with opposition also rising in markets such as Singapore and Malaysia. This sentiment is translating into real delays: some 168 early-stage US data center projects were facing cancellation or delay as of the start of August 2026, with around three-quarters of these delays linked to community opposition.

This resistance underscores a fundamental tension in the AI era. Companies want to deploy AI at scale to improve operations and sustainability, but communities are increasingly concerned about the environmental and infrastructure impacts of the data centers required to power these systems. Without stronger governance frameworks and transparent communication about energy use and climate impact, this opposition is likely to intensify.

The report suggests that the next phase of AI adoption will be defined not by technological capability, but by whether companies can expand responsibly. That means closing governance gaps, measuring climate impact rigorously, and engaging communities in decisions about where and how AI infrastructure is built. For now, most companies are still playing catch-up.