Egypt is stepping confidently into a regional leadership role by hosting the first edition of GITEX Global’s AI Everything in the country — a move that closely aligns with its national AI strategy and long-term economic ambitions.

Under its National Artificial Intelligence Strategy (2025–2030), Egypt is aiming to generate an estimated $42.7bn in AI-related economic value by the end of the decade. And with AI projected to inject up to $1tn into the wider regional economy, the country is making it clear it intends to be one of the key players shaping that growth story.

Held under the High Patronage of H.E. President Abdel Fattah El-Sisi, and hosted by the Ministry of Communications and Information Technology (MCIT) in strategic partnership with the Information Technology Industry Development Agency (ITIDA), AI Everything MEA is set to bring together leading AI enterprises, startups, policymakers, and thought leaders from more than 60 countries.

This year’s edition will feature over 350 companies, 250 AI startups and 100 investors managing a combined $50 billion in assets under management. Major global players, including IBM, Microsoft, Cisco, Cerebras, and Cloudera, are among those participating, alongside a growing wave of regional innovators.

Opening the Summit, Gitex’s global organizer, Trixie LohMirmand, also Executive Vice President, Dubai World Trade Centre and CEO, KAOUN International, says this inauguration doesn’t just mark the opening of a tech show, but “a symbolic opening of the new economic era for the region,” as she notes how AI is rapidly becoming the infrastructure of nations. 

LohMirmand says AI Everything is proud to contribute towards Egypt’s AI acceleration, from policies to implementation. She explains that as AI becomes the new operating system for countries, with new emerging competitors, suppliers, companies, policies, and regulations, this requires collaboration and convergence on a continental scale. 

“Egypt is known to be a builder. Historically, it has demonstrated its ability to create and globalize. So this is a moment for Egypt and the region to design systems that will reflect its language dominance, in ambition and markets.”

“This is the beginning and the launch of AI Everything Middle East and Africa in Egypt, and this shall be, I hope, the new National accelerated engine that will power Egypt and the region to lead, not just as a passenger.”

LANDMARK ANNOUNCEMENTS

Eng. Ahmed ElZaher, CEO of ITIDA, delivered remarks on behalf of Egypt’s Minister of Communications and Information Technology.

ElZaher emphasised that it is in Cairo, at the heart of the Middle East and Africa, that this new AI era must find its voice. “Today, we are adamant, as Egypt, to declare that our region will not only be present, but will shape, guide and lead — starting from here, from Cairo, Egypt,” he said.

He added that the inauguration of AI Everything reflects Egypt’s deep commitment to advancing the AI industry and its determination to position itself as the region’s leading platform for innovation, investment, and talent.

On behalf of the Ministry, ElZaher unveiled a series of new AI initiatives, platforms, and projects, beginning with the launch of Egypt’s first nationally developed, locally grounded large language model, named Karnak, after the Karnak Temple in Luxor.

Karnak is currently the highest-ranking Arabic LLM in the 30–40 billion parameter category, and also ranks competitively among models in the 70–80 billion parameter range.

“Karnak is going to be the foundation that we’ll proudly build upon to unlock further opportunities for startups and the private sector that view Egypt as a market for locally relevant AI solutions,” he said.

ElZaher also announced the rollout of two flagship applications. The first is a personalized AI tool designed to support education. At the same time, the second is an AI-powered assistant offering legal and regulatory guidance to help citizens and small businesses better navigate Egypt’s regulatory landscape.

AI is also being embedded into the Digital Egypt government services platform. Among the newly announced applications is Aqua, an NLP-based tool designed to audit calls between citizens and the Digital Egypt call center, flagging instances where incorrect information may have been provided or where service quality fell short. A second tool, React, is an LLM-based assistant built to enhance customer support operations.

In the healthcare sector, ElZaher highlighted locally developed AI engines trained on domestic data centers. These include solutions for the early detection of diabetic retinopathy and macular edema — conditions affecting people with diabetes — as well as AI-powered breast cancer detection systems.

“In addition, we are partnering with UNDP to organise these solutions and share know-how with our Arab and African neighbours through our AI Shared Initiative,” he added.

Further announcements included the rollout of automated translation services, an application called Top Command, an automated transcription engine, and another initiative, Be Al Masry, aimed at strengthening Arabic-language AI capabilities as part of Egypt’s positioning as a regional AI hub.

“Those tools will help citizens access information, opportunities, and services across language barriers, and will enable public institutions to serve more efficiently and more transparently,” ElZaher said.

At the Growth Stage, ITIDA also unveiled its newly redeveloped Egypt Innovate platform. El Zaher said the platform was designed to reflect and be built from within Egypt’s startup ecosystem. He noted that the platform aims to showcase innovations across the country’s entrepreneurship community and provide a clear, comprehensive view of the local ecosystem.

KEY PANELS

The summit’s first panel on the main stage discussed human-centered AI and featured Margaret Mitchell, Researcher & Chief Ethics Officer, Hugging Face, United States of America, Babak Hodjat, Chief AI Officer, Cognizant, United States of America, moderated by Mike Butcher, Founder & Editor, Pathfounders; Former Founding Editor-at-Large, TechCrunch.

Mitchell says that too often in technology development, teams begin by building a system they believe is impressive or technically advanced, and only afterward try to determine who it might serve. In this approach, she explains, technology is effectively retrofitted to human use cases.

A human-centric approach, she says, reverses that logic. Instead of starting with the technology, it begins by asking what people genuinely need, what matters most to them, and what their real priorities are.

From there, systems are designed to support those needs. Rather than focusing on applying technology to people, she says the goal should be to build technology that truly works for them.

Mitchell also touches on the intersection of AI and energy, saying this overlap demands more creative thinking. Even with state-of-the-art data centres, she says, there is significant room to improve efficiency — particularly by rethinking infrastructure and increasing reliance on renewable energy sources such as wind and solar.

She challenges the assumption that bigger is always better. “For a lot of tasks, you don’t need a massive model,” she notes, warning that large generative systems are often used where smaller, predictive models would suffice. “There are technologies that are much smaller that do the task just as well, if not better — and they require a lot less energy.”

Hodjat adds that there is progress on efficiency. “The good news is we’re getting smaller and more energy-efficient large language models as we go,” he says, noting that reasonably powerful models can now run on a laptop — something that was not possible just two years ago.

However, he cautions that rising adoption offsets those gains. “The more we use this stuff, the more we realize how useful they are — and so we use them even more. And they are very energy-hungry.”

His recommendation: invest in expanding processing capacity beyond large corporations. “If there’s one recommendation to make, it is investing in the availability of processing capacity outside of industry — to universities, to students.”

The panel is followed by a fireside chat with Dr. Ruchir Puri, Chief Scientist & Vice President, IBM Technology & Technical Community, IBM Research, who will delve into the practicalities of building sovereign AI. He explored critical questions around data ownership, computing infrastructure, and AI architectures — all key factors shaping long-term national and enterprise growth.

“You can partner with others or leverage external infrastructure, whether on-premises or in the cloud,” Puri said. “But true sovereignty comes from knowing which applications are running, ensuring compliance, and protecting your data. Data is the sovereign core — often the most critical and underappreciated element.”

He also cautioned against over-scaling by default, noting that many government and public-sector AI applications are better served by smaller, more efficient models, especially as energy consumption and operational costs become more significant.

A session featuring Amr Awadallah, AI Expert and Founder of Cloudera, and Omar Maher, Co-Founder and CEO of Monta AI, explored the challenges of taking large language models (LLMs) from pilot projects to full-scale production.

They highlighted that while LLMs are widely accessible, running them in real-world environments requires careful planning around data pipelines, inference costs, and architecture.

A fireside chat featuring Mirna Arif, General Manager, Middle East and Africa Growth Markets, Microsoft, focused on the complexities of deploying mission-critical AI across the Middle East and Africa, stressing that achieving AI sovereignty and trust by design is essential, especially in a region with diverse regulatory frameworks. 

A panel featuring Pastora Valero, Senior Vice President of International Government Affairs (EMEA & APJC) at CISCO, Amr Fathy, Chief Technology and Information Officer at e&, and Eng. Mahmoud Sofrata, Vice President for ICT Markets Development at ITIDA, examined how to add strategic value, operate AI systems at scale, and navigate the boundaries between national responsibility and private infrastructure.

Fathy explained that scaling AI requires a unified data layer across all companies and emphasized that AI infrastructure is a costly, ongoing commitment. “It’s a continuous investment,” he said. “It’s not just one capex with returns over 10 or 15 years — you need to refresh it on a yearly basis.” To build the necessary infrastructure, his team partnered with multiple providers to secure robust systems.

Sofrata, meanwhile, highlighted that scaling AI ultimately depends on citizen adoption. “AI is good, but for someone else, not me,” he noted, reflecting on a point raised earlier in the discussion. He stressed that widespread adoption depends on citizens at all levels of society seeing real value in AI, whether through government services, private-sector solutions, or financial institutions.

He added that, as other speakers had mentioned, while education and infrastructure are important, the ultimate measure of success is ensuring that these efforts lead to meaningful adoption by everyday users.

In a Fireside Chat on building trust in the age of intelligent finance, H.E. Dr. Mohamed Farid Saleh, Chairman of Egypt’s Financial Regulatory Authority, highlighted their efforts to develop responsible AI frameworks in financial services, balance innovation with market integrity, and explore regional regulatory harmonization in Africa.

Seizo Onoe, Director of the Telecommunication Standardization Bureau at the ITU, addressed the challenges of deploying public AI safely in a fireside chat, noting that rules often lag behind real-world applications and that global coordination is critical to ensure trust, durability, and inclusiveness in AI systems.

A panel titled “The Compute Choices That Define AI Power” brought together Dr. Haitham Hamza, Chairman at ITIDA-SECC and Dean of the Faculty of Computing and AI at Cairo University; Amr Elashmawi, Global Head and VP of Strategic Markets and Partnerships at Tenstorrent; Marc Zakher, Vice President & General Manager for the Middle East, Türkiye, and Africa at Cerebras; Nabil Badawy, Technology Director for Egypt, Sudan, and Libya at HPE; and Dr. Hoda Baraka, National AI Lead, Founding Director of ECRAI, and Advisor to the Minister at the Ministry of Communications and Information Technology, Egypt.

The discussion focused on how compute has become a real bottleneck as AI moves from pilot projects to full-scale production. The panel examined the risks of getting infrastructure decisions wrong — from locking in high costs and technical debt to creating long-term dependencies — and explored the hardware and system trade-offs required to run trillion-parameter models, including issues such as scarce GPUs, rising energy costs, and data center design. Speakers emphasized the importance of careful planning to ensure AI systems remain efficient, resilient, and scalable over the long term.

Discussing AI’s effect on energy, Golestan Radwan, Chief Digital Officer, United Nations Environment Programme,  Dr. Ahmed Tantawy, Founding Director, Applied Innovation Center, Egypt,  Mohamed Kassem, General Manager Microsoft, Egypt, and Dr. Hoda Baraka examined AI’s environmental footprint in Africa, highlighting its role in climate resilience, urban planning, and public services while stressing accountability, local control, and long-term sustainability.