Can AI help Syria recover from civil war?

1

The Pitch

How hackers are using AI

Good morning, everyone, and welcome to EnterpriseAM AI + Innovation! It’s a sunny early-fall morning in Cairo, where the team is still gathered after our EnterpriseAM Egypt Forum on Monday. More on that below. Let’s get started.

Isn’t it great when AI can take over a boring task for you? Yeah, hackers feel that way too.

Ye Olde AI Days: In the earliest days of AI, hackers might use the technology to help them craft a phishing email in a language they didn’t speak or to write a tricky piece of malicious computer code, Waseem Youssef (LinkedIn), Palo Alto Networks’ senior manager of systems engineering for the MENA region, told the audience at our Egypt Forum, which focused entirely on AI this year.

Now, hackers are handing the entire operation over to AI agents. And, just like legitimate businesses, some are achieving huge efficiencies. “You give it instructions and tell it, ‘I want to break into that company,’ and the agentic AI can actually go do reconnaissance,” Youssef said. “They can go and discover tools, try one way or another way, and then they will, at the end of the day, break in.”

Pwned in under an hour: The result is a massive decrease in the lag time between when attackers first target an organization and when they infiltrate it, per the 2026 Incident Response Report from Palo Alto’s Unit 42 division. The fastest 22% of hacks included in the report took under an hour, Palo Alto found, and the average time for the top quarter of attacks was 72 minutes. That’s down from roughly five hours in the previous year’s report.

What to do? Companies in the Arab world can follow some basic steps to protect themselves, panelists said: Invest more in cyber defense, make it a priority for the CEO and other top executives, and insist that everyone do the boring grunt work that makes companies better defended. That means updating software, ensuring company information can only be accessed by company employees, and canceling employees’ credentials as soon as they quit.

“Even if I have the top technology in my organization, I can still be penetrated if I’ve got a person who’s not able to use it wisely,” says Mohamed ElGarf (LinkedIn), Fortinet’s regional business development manager for Egypt and East Africa security operations, who spoke on the same panel.

But even after all of that, hackers enabled by AI — working harder and smarter than any human — will definitely have the advantage, says Sherif Shaltout (LinkedIn), managing executive for MENA at Liquid C2. “We always modeled security on the hacker mindset, which is that they would prioritize the highest value targets,” Shaltout says. “Now AI collapses all of that. They don’t have to make the choice between organization A and B. They can attack both and 10k more at [the same] time.”

Want more news and insights from our Egypt Forum on AI? There’s lots more in the Tuesday and Wednesday editions of EnterpriseAM Egypt — and still more to come in Forum Playback, which we’ll be sending out on Tuesday. –Joe

2

The Lede

Can AI help Syria recover from civil war?

A Saudi- and Silicon Valley-backed project in Syria is testing whether AI can help speed up recovery in a nation reeling from years of civil war. The project, led by the Silicon Valley firm Aixplain, has a dual mission: helping Syria’s telecommunications ministry build the basic infrastructure that will support AI projects and developing AI pilots.

Aixplain is working on two projects to start:

  • Software engineers at Syriatel, the nation’s largest telecom, are using an Aixplain-developed AI coding tool to develop new web and mobile services.
  • The company is also developing an Arabic-speaking AI agent for customer service assistance that will serve Syriatel’s more than 12 mn subscribers.

Why it matters: The project’s basic premise is that AI has so accelerated the pace of global innovation that Syria can’t afford to ignore the technology as it reenters the global economy.

In a nutshell: “If you’re planning to revamp the whole infrastructure in the country, AI is going to be critical,” Aixplain CEO Hassan Sawaf (LinkedIn) tells EnterpriseAM. “Whether you’re Syria or anyone else, it’s a mistake to not use AI in my opinion, because what you can do with AI is always faster. If you’re not using AI, you’re flying blind.”

Details:

  • Aixplain signed a memorandum of understanding with Syria’s Ministry of Communications and Information Technology in February, essentially to help the country develop AI infrastructure and services, including building infrastructure, securing data, and training Syrians in AI skills.
  • The partnership with Syriatel, finalized at the Leap Conference in Riyadh last month, is the first concrete and publicized deal under that MoU. Sawaf declined to describe the terms of the deal.
  • Aixplain is also working with Syrian private sector clients on AI projects, Sawaf says. He declined to name specific clients, but said the company is doing work with telecoms, banks, insurers, and in the education sector.

Roots in the country: Sawaf, whose parents emigrated from Syria to Germany in the 1960s, launched Aixplain in Silicon Valley in 2020 after stints at Amazon, Facebook AI, and eBay. The company also has a significant presence in Saudi Arabia.

The project highlights how AI differs from earlier technological revolutions — in theory at least. Two decades ago, when e-commerce was the hot new thing, urging a shattered country like Afghanistan to build an e-commerce sector would have made little sense without functioning banks and payment systems. (Syria, too, is struggling with those.) AI tools, by contrast, are far more accessible and the cost savings typically come much quicker.

“It’s a lot more intuitive for the average citizen,” Syriatel CEO Yasir Kahf (LinkedIn) tells EnterpriseAM.

Assessing the challenge: One big challenge Syria is facing is that the years its businesses lost to the civil war were also years that saw huge technological advances in the rest of the world, Kahf says. That makes anything Syriatel can do to speed up its modernization vital, he says.

“Instead of trying to solve today’s problems with yesterday’s tech, we’re trying to solve it with tomorrow’s tech,” he says. “These are projects that in a traditional programming environment would have been a one-and-a-half, two-year, three-year project, and we're now getting them done in two or three months.”

Not overambitious: “If they were going to run elections and they’d never had free elections before but now they want to turn it all over to AI, I would say, ‘There’s no way in hell that’s going to work.’ But they’re not being too ambitious. I would say the risk is much smaller,” Kentaro Toyama (LinkedIn), a University of Michigan professor who studies technology and international development and a well-known critic of overstating the benefits of new technology, tells EnterpriseAM.

BACKGROUND- Syria emerged from more than a decade of devastating civil war in December 2024 with the sudden downfall of its leader, Bashar Al Assad. The US began lifting crippling economic sanctions on the nation in 2025, after which the new government, led by President Ahmed Al Sharaa, began courting international investors. The World Bank puts the cost of Syrian reconstruction at a jaw-dropping USD 216 bn — that’s about 10x Syria’s GDP.

To be sure, there are significant infrastructure barriers to broad AI adoption in Syria. The country has only a limited amount of fiber-optic cable carrying internet traffic, though that will increase with the deployment of SilkLink, a USD 800 mn joint Saudi-Syrian project that aims to build a 4.5k-kilometer fiber-optic network across Syria and link it with Europe and Asia.

The country is also short on GPUs, the high-powered computer chips that power most AI work. Current AI deployments rely, at least partly, on GPUs in Saudi Arabia and Europe, Sawaf says. The nation has made “data embassy deals” with those countries so it retains legal control over data processed there, he says.

To cap it all off, the country is short on energy. Electricity availability in most provinces is now between 18 and 20 hours per day, up from just a few hours per day previously, Energy Minister Mohammad Al Bashir said last month, per state news agency Sana.

“Infrastructure doesn’t exist like we’re used to in the West and that’s a charter we have, to help them bring their technology and their processes to where they should be in 2026,” Sawaf says. Many Syrian companies are also experimenting with cloud-based AI tools hosted by Amazon and Microsoft, he says.

Analysts consulted by EnterpriseAM generally praised the Syrian AI pilots for not being overambitious and focusing on easy cost savings. But they also warned that relying too much on AI could backfire — and that deploying AI or other technology can’t substitute for the hard work of building the government and private sector institutions that Syria will need to build a vibrant economy.

The cemetery of tech failures: “There’s a cemetery's worth of projects where somebody thought, ‘Oh, I’ll bring some digital technology in and it will solve this government service problem or this social problem or this political problem.’ And technology is never the thing that changes those kinds of things,” Toyama says.

Rules of the road: In addition to building tech infrastructure, the Syrian government should focus on ensuring transparency around how AI is used and governed and on creating strong legal rules around AI, experts say. That will help reduce international concerns that AI investments now could be used to fund a future surveillance apparatus, they add.

The government should also be thoughtful about the way AI implementations could contribute to rising unemployment. That’s a concern across the globe, but further unemployment could be extra destabilizing in a nation that’s still trying to get its footing after so many years of crisis. Syria’s youth unemployment rate was about 33% in 2025, according to World Bank data. Analysts often look at youth unemployment as a key measure of the risk of unrest in frontier markets.

“The investment in digital infrastructure in this area can have huge beneficial impacts for Syria if done correctly,” Nadia Farra (LinkedIn), a former US government official focused on the Middle East and technology and a nonresident senior fellow at the Atlantic Council, tells EnterpriseAM. “And by correctly, I mean all the really boring, unsexy things you have to do to make sustainable economic growth happen.”

3

The Business of AI

How would an AI development slowdown affect MENA+?

Would the proposed AI slowdown that’s pitting Silicon Valley against Washington hurt the Gulf’s AI ambitions? Depends on who you ask.

Some analysts see real danger. This argument basically holds that Gulf AI development is inextricably linked to the health of the US AI labs including OpenAI and Anthropic — which has been growing faster than basically any industry in history up until now. If it contracts, the Gulf will feel it, they say. TL;DR: When the US AI industry gets a cold, the Gulf gets pneumonia.

“The AI growth boom is intrinsically tied now to economic horizons for Gulf states,” Nick Shafer (LinkedIn), an emerging markets strategist focused on MENA and technology and an adjunct professor at the University of California-Berkeley, tells EnterpriseAM. “That means that whatever happens in San Francisco will ripple into the growth models for these countries.”

Then there’s the rosier view — essentially that the Gulf AI economy right now is mostly about building infrastructure — and that the demand for AI infrastructure in the region won’t dip just because model improvements aren’t super jaw-dropping for a while.

Pay attention to the infrastructure: “My takeaway for the Gulf / MENA [is that] the AI models themselves are important, but pay attention to the infrastructure buildouts,” Vincent Carchidi (LinkedIn), a defense industry analyst at Forecast International, tells EnterpriseAM.

What kind of slowdown? The slowdown model that gripped the Washington narrative last month is unlikely to happen. That model — proposed by Anthropic CEO Dario Amodei and embraced by other AI leaders — envisioned a coordinated pause by AI labs while human data scientists and outside auditors worked on better understanding why their models were going haywire and hacking things they weren’t supposed to. But with President Trump firmly not on board, such collective action is probably impossible.

But there’s some evidence AI firms are tapping the brakes independently. OpenAI canceled the release of its most recent model, GPT-6.1 Astra, over safety concerns last week.

There’s also a risk that hyperscalers will grow increasingly gun-shy — and slow their release schedule — as the legal liability from rogue agents racks up. Legislators in the US are also pushing to ensure AI companies bear the legal burden of their agents’ bad deeds.

The popping bubble scenario: There’s also a possibility that investor confidence in AI could sag. That could hurt the Gulf both as a customer of AI innovations coming out of the US — and as an investor in many of the top AI companies, where sovereign institutions from Saudi Arabia, the UAE, and Qatar have taken positions in labs including xAI, OpenAI, and Anthropic.

The story in Egypt is a little clearer. The big story there is still mostly about getting companies to use AI to create efficiencies and save money — not about doing anything at the bleeding edge, executives told us on the sidelines of our Egypt Forum in Cairo this week. The forum focused this year exclusively on AI.

A different lens: “We see that global narrative on why development of these models needs to slow down until you have the right governance and measures in place,” Bassem Fayek (LinkedIn), managing director and partner at Boston Consulting Group in Cairo, tells us. “In Egypt, we are far from thinking with the same lens. We have a much more important lens to look through, which is: How are we able to use such tools and capabilities to make an impact?”

4

AI in Business

TII launches an AI model built for Emirati Arabic

Abu Dhabi's Technology Innovation Institute (TII) launched its new Falcon-Emirati model this week, trained specifically on the Emirati dialect of Arabic. The 7-bn parameter model is part of a push to make Arabic-language models more useful by training them on idioms and slang rather than formal language.

Why it matters: Arabic AI models have historically suffered from limited data, so many models were built on English data translated into Arabic, Dr. Hakim Hacid (LinkedIn), chief researcher at TII’s AI and Digital Research Center, tells EnterpriseAM. That approach “resulted in losing all the specificities of the languages, of the culture,” he says. TII also released a multilingual speech-transcription model and an Arabic model for extracting text from images and documents.

How it scores: Falcon-Emirati scored 84.83% on an Emirati Arabic test, beating every open-source Arabic and multilingual model it was compared with, TII says. The institute designed the test itself, but also made its data public so outsiders can check the results. Closed models like ChatGPT and Claude weren’t part of the comparison.

Next: TII plans to build models for other Arabic dialects, but wants user feedback on the Emirati model first, Hacid says. He declined to give a timeline for when other models will arrive.

Other dialect models: Teams led by Mohamed bin Zayed University of Artificial Intelligence, Abu Dhabi’s AI research university, released Atlas-Chat, a model for Moroccan Arabic, in 2024, and Nile-Chat, for Egyptian Arabic, last year.

5

Dealbook

OpenAI is seeking more UAE funding

OpenAI goes back to the Abu Dhabi well: OpenAI is in talks with UAE funds to provide up to USD 10 bn to help anchor its upcoming USD 30 bn funding round, Bloomberg reports. Once again, MGX, the AI investment firm set up in 2024 by sovereign wealth fund Mubadala and AI champion G42, is at the table.

MGX first backed the ChatGPT maker in its USD 6.6 bn funding round in 2024. It then co-led the company’s record USD 122 bn round this March. (MGX also backs rivals Anthropic and xAI, and raised a USD 49 bn AI fund in July.)

New financing for Broadcom chips: Broadcom has started lining up a USD 60 bn debt package to help fund the AI chips it’s building for Anthropic and other firms, Bloomberg reports. The deal is a test of whether investors are still keen to fund AI’s US buildout amid a public backlash against data centers.

Meanwhile: SpaceX is looking to borrow USD 40 bn to buy Nvidia chips, per MarketWatch.

Finally: Anthropic’s IPO is still tentatively on for November — on month than originally planned — according to the Wall Street Journal.

6

The Security Beat

OpenAI agents went rogue again — targeting Wikimedia this time

Another day, another rogue agent attack: Rogue OpenAI agents made unauthorized edits and fired millions of automated requests at the servers of Wikimedia, the parent of the online encyclopedia Wikipedia, the organization said this week.

Assessing the damage: Wikimedia found no evidence that its systems or data had been compromised. Most edits were in wiki testing areas rather than pages visible to general readers, although some changes to a citation tool were potentially malicious. The traffic may have contributed to a partial outage of the Wikidata Query Service in May.

It’s the pattern that matters, more than any individual rogue action. The reports of rogue agent activity have been tumbling in roughly twice a week (according to trackers) since the Hugging Face incident in July, which captured the world’s attention and prompted AI leaders to call for a development slowdown.

Data point: The New York City Council summoned representatives from top AI companies for a hearing this week, underscoring the nationwide concerns about the technology. In one notable exchange, Council speaker Julie Menin slammed company officials for saying they couldn’t quantify the risk of a “catastrophic scenario” caused by AI and insisted they should not release models they can’t control. Here’s more from The New York Times.

Who gets to ask questions: Note that the New York City Council is one of the very few government bodies with enough clout to force representatives from the big AI labs to show up and answer questions — a reflection of the largely oversight-free universe in which OpenAI and its competitors exist. (SpaceX, run by Elon Musk, refused to send a representative to the hearing despite receiving a subpoena, legally compelling the company send a representative.)

7

The Global Beat

The US still fears an AI bubble

Is AI a bubble about to burst? Not a chance, US Treasury Secretary Scott Bessent says, per Bloomberg. He cited big spending by the likes of Microsoft, Google, and Meta as well as revenue growth at Anthropic and OpenAI.

And yet: Ray Dalio, founder of Bridgewater, the storied US investment firm, called AI a “classic bubble” nearing its bursting point, Bloomberg also reports. Dalio cited rising interest rates and the debt piling up to fund AI infrastructure.

The question is nearly existential for the Trump administration, which some analyst think faces bleak prospects in the upcoming November midterm elections. AI is just about the only growth sector in the US economy right now and any downturn could hit both financial markets and the real economy in ways that tank Trump’s already-abysmal approval ratings.

As we noted above: Infrastructure and other investors in the Gulf are keeping a close eye on chatter that there’s a bubble that could burst.

The good news: US software stocks hit their highest level since November 2025 on Tuesday, Reuters reports, suggesting some investors don’t see there’s a bubble.

8

Building Blocks

Khazna covers 28% of its cooling electricity with carbon-free power

G42's data center arm Khazna is using renewable energy for about 28% of its cooling needs, according to the company’s first environmental and government report out this week.

Big plans: Khazna is aiming for 50% renewable energy by 2030 and 100% by 2035. (Those figures refer specifically to the electricity for cooling, not the electricity that powers the servers themselves.)

Why it matters: Khazna controlled more than 70% of the UAE's data center capacity as of last year, Rest of World reported, and it’s leading the buildout of Stargate UAE, the 1 GW AI cluster G42 is developing with OpenAI, Oracle, and Nvidia.

9

The Chaser

OpenAI solves more math problems and there’s a new French model

Proof positive: OpenAI says its AI has solved 377 previously unsolved math problems, The New York Times reports, and has released the proofs for scrutiny. The news comes one month after the company solved a math problem so hard it came with a USD 1 mn prize.

Le Chonk est arrivé: France's AI startup Mistral unveiled its new large language model in Abu Dhabi on Tuesday. It’s nicknamed “Le Chonk,” but with the much more boring official name Mistral Large 4. The company says it beats some Chinese models on cybersecurity. Here’s more from Reuters.

The French have a talent for getting really weird on film. Perhaps the country’s most famous director, Jean-Luc Godard, was nearly incomprehensible for a big chunk of his career. Try 1967’s Weekend for a look at when he was still more weird-entertaining than just weird-weird. For totally sublime weirdness, try Leos Carax’s 2012 film Holy Motors, but take our advice and go in blind. No Googling.

That’s it for this week! Like what you’re reading? Feel free to invite some friends. Or drop us a line at [email protected]. We’ll be back in your inbox on Tuesday.


October 2026

11-13 October (Sunday-Tuesday): Riyadh’s King Salman Center for Disability Research runs an AI hackathon for people with disabilities, sponsored by Boeing Saudi Arabia.

26-27 October (Monday-Tuesday): The Dubai AI Festival returns as part of Dubai AI Week.

26-29 October (Monday-Thursday): Riyadh hosts its annual Future Investment Initiative under the theme “The Power of Legacy” with a heavy AI-investment track.

27 October (Tuesday): Alphabet reports 3Q earnings after the US market closes, with AI capex and Gemini / cloud growth as the headline watch items.

28 October (Wednesday): Microsoft reports fiscal 1Q 2027 earnings after the US market closes, including Azure and OpenAI-partnership capex disclosures.

28 October (Wednesday): Meta reports 3Q earnings, with AI infrastructure spend in focus.

29 October (Thursday): Amazon reports 3Q earnings, including AWS and Bedrock AI growth.

30 October (Friday): UAE businesses with annual revenue of AED 50 mn or more must have an accredited e-invoicing provider in place by this date.

November 2026

7-10 November (Saturday-Tuesday): Cairo ICT runs in New Cairo alongside AIDC Expo, its dedicated AI, data center, and cloud event.

17-19 November (Tuesday-Thursday): Bengaluru Tech Summit brings businesses, startups, researchers, and policymakers to Bengaluru, India, under the theme “AI & Beyond.”

17-20 November (Tuesday-Friday): Microsoft Ignite runs in San Francisco and online, covering Microsoft’s AI, cloud, security, and data products.

18-19 November (Wednesday-Thursday): Global AI Show opens at Space42 Arena in Abu Dhabi, with enterprise AI, computing capacity, and sovereign AI on the agenda.

30 November-4 December (Monday-Friday): AWS re:Invent brings Amazon’s cloud and AI conference to Las Vegas, with product announcements and sessions on deployment, costs, and security.

December 2026

1-3 December (Tuesday-Thursday): Black Hat MEA brings cybersecurity leaders, researchers, and technology companies to Malham, Riyadh.

6-12 December (Sunday-Saturday): NeurIPS holds its main AI research conference in Sydney, offering a look at advances that could shape the next generation of commercial AI.

7-11 December (Monday-Friday): GITEX Global opens with a leadership summit at Dubai World Trade Centre on 7 December, followed by its exhibition at Expo City Dubai on 8-11 December.

Now Playing
Now Playing
00:00
00:00