The Middle East and North Africa ranks among the world's leading regions in Gemini adoption, according to Google. Within businesses, however, the picture is more complex. Most organizations are now experimenting with artificial intelligence or running pilot projects, but moving from limited trials to large-scale deployment remains challenging, particularly when it comes to data and governance.
This contrast between rapid consumer adoption and the challenges of enterprise implementation presents the region with a different test as 2030 approaches.
Anthony Nakache, Managing Director of Google for the Middle East and North Africa, explained in an exclusive interview with Asharq Al-Awsat that the next phase will not be measured solely by user numbers or infrastructure spending. Instead, success will depend on the region's ability to develop homegrown technologies and export them to global markets, rather than simply adapting solutions developed elsewhere.
The Production Test
Nakache, who has spent 18 years at Google and approximately four and a half years in the region, believes the Middle East has become a major player in the global AI landscape.
He attributes this position to a combination of factors, including government ambition, national strategies and infrastructure investment, alongside skills development, policymaking, and support for scientific research and startups.
The results of these efforts are already evident in adoption rates, particularly among consumers. The region ranks among the world's leading markets for Gemini usage, accompanied by high levels of optimism about AI's potential.
However, Nakache believes the next stage will require broader commercial adoption within organizations, eventually extending to more advanced systems, including agentic AI, which can execute multiple tasks and address increasingly complex problems.
He places trust at the center of this transition, particularly as AI systems evolve from providing answers to taking actions. He also emphasizes the importance of developing solutions that reflect local contexts alongside global technologies.
Impact Beyond Technology
Nakache points to the economic impact of Google's products in Saudi Arabia and the United Arab Emirates to illustrate how the company measures the value of technology beyond its direct services.
According to a study conducted by an independent consultancy commissioned by Google, the economic impact of the company's products across the two countries was estimated at approximately $14.2 billion in 2024.
These products include YouTube, Maps, Search, advertising, cloud computing and Android, all of which incorporate AI to varying degrees.
Nakache links this economic value to the ways individuals and businesses use these tools to search for information, make decisions, acquire skills, reach customers, and improve efficiency and security.
He also highlights employment opportunities generated by the Android and cloud computing ecosystems and YouTube's creator economy.
In his view, technological leadership is not limited to developing AI models and infrastructure. It also encompasses measurable economic and social impact.
Why Do Some Companies Struggle?
At the enterprise level, Nakache does not consider insufficient infrastructure or a lack of tools to be the only obstacles to scaling AI, although he anticipates that additional capacity will be needed to meet future demand.
The more pressing challenge, he explains, is moving from pilot projects to full implementation.
He rarely encounters an organization that is not experimenting with AI in some form. Yet experimentation does not necessarily mean that the technology has become an integral part of everyday operations.
He identifies two principal obstacles.
The first concerns data. Organizations need to understand what data they possess, how accurate and up to date it is, and how AI systems rely on that information when making decisions, even as newer models become increasingly capable of handling less structured data.
The second challenge is governance, particularly determining accountability for decisions made by agentic systems.
Nakache notes that relatively few companies he works with have established a clear framework for addressing this issue, making the transition from controlled testing environments to broader customer-facing applications more complicated.
Challenges of Practical Adoption
While data and governance are major challenges for businesses, consumers face a different difficulty: learning to use AI tools effectively.
Nakache explains that many users still struggle to obtain their desired results because the prompts they provide are too limited or unclear.
This challenge becomes more pronounced with the emergence of agentic platforms capable of automating tasks, as users increasingly need to understand how to design workflows rather than simply formulate questions.
For this reason, Google is seeking to integrate AI into products that individuals and businesses already use, rather than consistently presenting it as a separate tool.
Nakache cites Performance Max, Google's AI-powered advertising solution, as an example. The platform allows advertisers to provide creative assets and business objectives while the system manages campaign delivery across Google's products.
According to Nakache, Google has observed sales increases ranging from 14% to 20% among users of the tool in the region.
The Skills Gap
Nakache believes the digital skills gap is not a new problem, but AI has made it more pressing because of the technology's advanced capabilities, associated risks, and the need for skills extending beyond traditional digital literacy.
Google says it has trained more than three million people across the region in digital skills since launching its Maharat min Google program in 2018.
The company has also committed $15 million through its AI Opportunity Initiative for the Middle East and North Africa, which aims to equip half a million people with AI skills in cooperation with local partners, including government entities, universities and business incubators.
Nakache stresses, however, that success should not be assessed solely by the number of trainees or certificates awarded. It must also reflect participants' ability to apply what they have learned in practical settings.
Accordingly, the initiative targets a diverse range of participants, including students, government employees and entrepreneurs.
Despite the region's high AI adoption rates, Nakache believes the more significant gap remains in the capabilities required to build technologies and integrate them into organizations, although some countries have succeeded in attracting international talent.
Changing Search Behavior
AI's influence also extends to Search, one of Google's oldest and largest products.
Nakache says 15% of daily searches involve queries the company has never encountered before, reflecting the continued evolution of user behavior.
As AI becomes increasingly integrated into the search experience, Google's AI Overviews feature has reached approximately 2.5 billion monthly active users, while AI Mode has reached one billion users, according to company figures.
The nature of search queries is changing as well.
Questions submitted through AI Mode are now approximately three times longer on average than conventional search queries, while more than one in six queries go beyond text to include images or other forms of interaction.
Combating Digital Fraud
Nakache acknowledges that AI has transformed the digital threat landscape, although online fraud existed long before the technology emerged.
He emphasizes, however, that Google also uses AI as one of its principal defenses against malicious activity.
According to figures he provided, the company's technologies helped block or remove approximately eight billion harmful advertisements over the past year, along with nearly 25 million accounts associated with malicious actors.
To address synthetic content, Google developed SynthID, a technology recognized by TIME magazine as one of the Best Inventions of 2026.
SynthID embeds imperceptible digital watermarks into AI-generated text, images, videos and audio without compromising their quality.
According to Google, the technology has been used to watermark more than 180 billion images and videos, as well as the equivalent of 240,000 years of audio content.
The company's content verification tools also handle approximately one million requests daily, while the technology's use is expanding to include companies such as OpenAI, NVIDIA, Kakao and Apple.
These tools help verify the presence of digital watermarks in content supported by the technology, but they do not guarantee the detection of all AI-generated material.
Nakache emphasizes that combating fraud and synthetic content cannot be the responsibility of a single company. Instead, it requires cooperation among governments, regulators, academic institutions and the private sector.
The 2030 Test
As 2030 approaches, Nakache believes the region's success in AI will be measured not only by adoption rates and investment levels, but also by its ability to develop proprietary technologies and bring them to global markets.
He considers attracting talent one of the most important factors in the transition from technology consumption to production.
Saudi Arabia, the UAE and Qatar are already developing Arabic-language models, he notes, but these efforts represent only the beginning of what the region could achieve.
The presence of Arabic speakers beyond the region's borders also provides a wider potential market for locally developed solutions.
Meanwhile, the combination of infrastructure, supportive policies, investment and skills development could enable the creation of technologies with global reach.
For Nakache, the real test by 2030 is whether the Middle East and North Africa can move beyond using technologies developed elsewhere or adapting them to local needs, and instead produce innovations that originate in its own markets and reach users around the world.
