08 Jun 2026

Okta building a proactive strategy for sustainable AI

This guest blog was written by Alison Colwell and Sophia Gluck, from Okta.

Around the globe, artificial intelligence (AI) adoption has surged over the last year. In the UK, 70% of organizations have begun leveraging AI capabilities, and the potential for increased efficiency and productivity is sparking new implementations across business functions.

With a rapidly evolving technology like AI, and the constant industry chatter about its use cases, organizations often feel a sense of pressure to move quickly. When it comes to AI adoption, leaders may be more focused on keeping pace with industry peers, but sustainability is quickly becoming a key consideration as companies roll out more AI-powered tools.

AI stands to represent a significant environmental footprint. In 2026, the International Energy Agency (IEA) has forecasted that AI-driven data center demands to increase overall data center energy consumption by 20 percentage points. We believe that organizations should begin working to understand and manage their usage of AI sustainably.

Okta’s sustainable AI strategy

Since 2020, Okta has been on a journey to understand and address its climate impact, a commitment that has deepened with time. As part of our journey, we have expanded our annual greenhouse gas emissions to cover all relevant scopes; expanding our 100% renewable electricity program to cover remote work and third-party cloud services, as well as our global offices; and expanding focus areas to sustainable finance and sustainable AI.

With the technology industry, and other stakeholders such as customers and investors, increasingly focused on AI adoption, Okta is investing more heavily in AI to drive innovation and productivity. Okta is committed to helping organizations secure AI while also developing a strategy for sustainable AI.

We are developing our Sustainable AI strategy to align with our climate goals: measure, reduce, invest in renewable electricity, and engage our vendors.

Measure and manage

Assessing AI’s environmental impact is challenging due to the lack of a single, industry-wide standard for measuring and reporting the energy and carbon emissions of AI tools and models. The other challenge is the lack of disclosure and tools from the AI industry on reporting their customer’s emissions. In this rapidly evolving landscape, Okta developed an interim methodology to help us navigate this new terrain. This approach is built on publicly available research and data, which may help Okta to estimate the environmental impact of our AI usage today. We will continue to advocate for greater transparency and standardization from our partners and the industry at large.

We first identify a use case, such as AI-assisted code generation. Then we apply a three-step process as follows:

Measure Usage Activity: we look at our own usage data, such as the number of lines of code generated and the volume of chat interactions.
Convert Activity to Energy Consumption: we translate this usage activity into energy consumption. This involves converting lines of code and chat messages into "tokens" (the units of text that AI models process) then using energy profiles for different AI models to estimate token energy density. We also include energy consumption from the model training that can be associated with Okta then using the trained model.
Calculate Carbon Emissions: we calculate the carbon footprint based on the estimated energy consumption and the data center locations. We consider employee distribution to map cloud data center regions and associated carbon intensity for inference. For model training, we estimate data center location based on the AI vendor and the AI model. This includes the operational emissions from the electricity used to power the AI models, the embodied emissions associated with manufacturing the server hardware, and training emissions attributed to Okta.

We believe this methodology, although reliant on a number of assumptions and subject to uncertainties, is a step forward in helping us to better understand our impact and providing a baseline from which we can improve. It spans from both limited specific model architecture disclosure for closed models, as well as data center infrastructure running these models. We will be using this information for internal purposes, and will not be integrating it into our greenhouse gas inventory. This will serve as an initial understanding of our AI environmental footprint, allowing us to identify ways to invest in decarbonization.

Reduce

Our climate strategy prioritizes reduction, or using resources efficiently to save energy and money. As we further develop and implement our Sustainable AI strategy, we created a resource for employees as a way to deepen their understanding of sustainable AI use. The three guiding principles of this framework include:

using AI for the greatest impact;
choosing the appropriate/lightest model for the use case; and
prompting efficiently, such as short and focused conversations, providing the model with only relevant reference documents and telling the model where to look for context, and providing examples.

Purchase renewable electricity

We aim to expand our 100% renewable electricity program to purchase renewable electricity to match our electricity consumption from our top AI enterprise tools. Our annual renewable electricity program currently matches electricity consumption from our global offices, remote workforce, and third-party cloud services. As part of our renewable energy program, we prioritize both reducing emissions and providing additional societal impacts such as energy access, public education, local jobs and skill-based training, economic development, and community-based healthcare and education.

Engage vendors

In 2025, we focused on engaging with our cloud service providers and AI vendors to leverage our purchasing power to push for more sustainable industry-wide actions related to sustainable AI. We also provided sustainability questions to our procurement team and business units that manage these vendor relationships to encourage them to incorporate sustainable AI into their discussions.

What’s next - working collaboratively

The rapid adoption of AI necessitates a proactive, flexible, and comprehensive strategy to manage its environmental impact. Okta's four-pillar approach—Measure and Manage, Reduce, Purchase Renewable Electricity, and Engage Vendors—provides a framework for embedding sustainability into our business operations. While challenges like industry-wide measurement standards persist, the collective effort from organizations to measure their impact, prioritize reduction, match energy use with renewable sources, and leverage purchasing power to hold vendors accountable is crucial to ensuring sustainable deployment of AI. We believe a path to a sustainable future for Cloud, Data, and AI includes continued transparency and collaboration across the industry.