1. Many organizations in the Kingdom have already run AI pilots, but far fewer have moved them into daily operations. What is holding that transition back, and what changes once the right tooling is in place?
That is a very good question. What we are seeing right now is not unique to Saudi Arabia; it is a global challenge.
First, the AI landscape is changing extremely quickly. Tools and models evolve almost daily, and a model that is considered leading today may be surpassed tomorrow. Second, organizations face an overwhelming number of options, making it difficult to determine which technologies are most suitable for their specific needs. The third challenge is that the return on investment from agentic AI has not yet been fully proven across all use cases.
That said, Saudi Arabia is moving very quickly when it comes to AI adoption. The Kingdom is among the global leaders in this area, with a significant number of organizations already using AI in their day-to-day operations.
The next stage, in our view, is what I would call secure productivity.
In software development, for example, organizations need to continue giving developers the freedom to use the AI tools and agents that help them work effectively. At the same time, business leaders and CTOs are increasingly asking important questions: How much is this costing us? What is the quality of the output? Is it secure?
This is why we believe organizations are increasingly moving towards stronger governance and control over AI usage. They need visibility into how AI is being used across teams, how usage is allocated, which models are permitted, and how security and compliance requirements are being managed.
The requirements will naturally vary between industries. A bank, for example, is likely to have much stricter controls than other organizations because of regulations, data protection requirements and other compliance considerations.
Organizations also need to understand the return on their investment: how much they are spending, how resources are being allocated, and whether they are achieving the expected quality and productivity gains.
This makes governance, analytics and auditing increasingly important.
Do you think this transition will be a slow process?
I think the market is moving so quickly that, in some cases, organizational processes are struggling to keep up. The technology is evolving at a pace that traditional decision-making and governance structures were not necessarily designed for.
2. Agentic AI can act, not just advise. What does a governance framework need to cover to give these systems real autonomy without losing control or accountability?
That is another very important question.
At JetBrains, we believe there is no denying that AI is here, and it is becoming an integral part of software development. AI agents have made it incredibly fast and inexpensive to generate code. Generating code itself is no longer the main challenge.
The real question is control and accountability. Can you trust the code? Can you confidently deploy it into production? Ultimately, a machine cannot take responsibility for the final product.
We believe there are two important elements.
First, human accountability has to remain clear. Developers should manage AI, not the other way around. Developers review the code, validate the output and ultimately decide what is deployed into production. The machine is there to assist them, but the responsibility remains with the developer.
That is a fundamental part of how we approach AI at JetBrains.
The second element is organizational governance. Companies need processes and frameworks covering security, costs, auditing and the way AI is used across the organization.
Each organization is ultimately responsible for establishing its own governance framework based on its specific business, regulatory and security requirements. Our role is to support that process through capabilities such as analytics and our open agentic system.
3. As organizations move from AI experimentation to wider adoption, what skills, infrastructure and organizational changes are most critical to ensuring AI delivers measurable business impact?
From our perspective, particularly in the software development space, there are three key aspects to delivering measurable business impact.
The first is cost efficiency.
AI can be extremely powerful, but if it is used inefficiently, it can also consume a significant number of tokens and become expensive. We are moving beyond simply looking at software licence consumption towards understanding the overall cost of AI usage.
From a developer’s perspective, it is important to use professional tools with the intelligence and context needed to help AI agents generate better code, with fewer errors and less unnecessary token consumption.
The second aspect is quality.
If an AI agent operates within a professional development environment, such as JetBrains IDEs, it can benefit from deep intelligence and a strong understanding of the context of the code. JetBrains has more than 25 years of experience and intelligence built into its development tools.
This enables developers to produce code that is not only faster to generate, but also potentially more secure and of higher quality.
The third aspect is latency and speed.
Ultimately, organizations need to consider all these factors together: how quickly they can achieve results, how good those results are, how secure they are, and how efficiently they are produced.
For us, measurable business impact comes down to finding the right balance between cost, quality and speed
Source: Riyadh Daily