Talabat Holding, the on-demand online ordering and delivery platform in the MENA region, has reported robust pro forma financial results for the three-month and six-month period ended 30 June 2025, supported by double-digit growth across GCC and non-GCC markets.

Gross merchandise value (GMV) rose 32 per cent year-on-year to $2.4bn, or 33 per cent on a constant currency basis. Revenue increased 35 per cent to $982m (36 per cent constant currency), while adjusted EBITDA grew 31 per cent to $166m, representing 6.8 per cent of GMV. Net income climbed 33 per cent to $119 million, or 4.9 per cent of GMV. On a normalised basis, adjusting for non-recurring items, net income grew 25 per cent to $116m.

The performance was driven by strong demand in both the Food and Grocery & Retail (G&R) verticals, with growth reflecting accelerated customer acquisition and increased order frequency. Ramadan’s first-quarter impact, compared with the previous year, also contributed to the results.

Read: Here’s how customers can benefit from talabat and Bolt’s new partnership

Given the momentum, talabat has revised its 2025 full-year guidance upwards. GMV growth is now forecast at 27–29 per cent on a constant currency basis (previously 17–18 per cent), revenue growth at 29–32 per cent (previously 18–20 per cent), adjusted EBITDA margin at 6.5 per cent (unchanged), net income margin at 5.0 per cent (unchanged), and adjusted free cash flow at 6.0 per cent (unchanged).

Other highlights included strong adoption of talabat pro, its subscription loyalty programme, and a shift in GMV product mix that, despite lower gross profit margins, was offset by improved cost margins. GCC markets accounted for 83 per cent of GMV, with non-GCC markets representing 17 per cent.

Tomaso Rodriguez, chief executive officer of Talabat, said: “We have achieved another strong quarter of financial and operational results, fueled by significant customer acquisition and increased order frequency. Our ongoing commitment to enhancing the consumer value proposition, expanding our Groceries and Retail vertical and fostering deeper customer loyalty is clearly yielding results. We are particularly pleased with the strong uptake of talabat pro, our premium subscription loyalty programme, across all markets, alongside strong growth in demand within our non-GCC markets.

“This growth complements the continued strength of our core GCC markets and the strong performance of our Food vertical. The UAE, our largest market, maintained its robust growth trajectory in line with the overall pace of the Group. Kuwait, our most established market, delivered impressive growth of over 20 per cent for both the quarter and the first half of the year. Likewise, our Food vertical grew more than 20 per cent year-on-year, reinforcing its strong contribution to our overall growth. With this momentum, we are confident in our outlook and are pleased to raise our full-year guidance across all metrics.”

The full set of disclosures can be found within the Investor Relations section on talabat’s website.

Abu Dhabi’s artificial intelligence (AI) sector has expanded 61 per cent between June 2023 and June 2024, reaching 673 companies and positioning the emirate as the fastest-growing AI hub in the Middle East and North Africa (MENA), data from the Abu Dhabi Chamber of Commerce and Industry (Abu Dhabi Chamber) showed.

The emirate now accounts for a notable share of the world’s 90,904 AI companies and is aiming to cement its role as a global leader in AI-driven enterprise, innovation and research.

The growth is underpinned by institutions such as the Mohamed bin Zayed University of Artificial Intelligence (MBZUAI) – the world’s first graduate-level research university dedicated to AI – which ranks among the top global institutions for AI research impact.

The university’s programmes in machine learning, computer vision, natural language processing, robotics and data science are drawing top-tier talent.

AI ecosystem supported by AIATC, ATRC, TII and others

Abu Dhabi’s AI ecosystem is also supported by the Artificial Intelligence and Advanced Technology Council (AIATC), the Advanced Technology Research Council (ATRC), the Technology Innovation Institute (TII), AI71, Hub71, G42 and Space42.

These bodies are advancing adoption in sectors from healthcare and logistics to finance and education, with TII and MBZUAI developing open-source AI models accessible to researchers worldwide.

More than 58 per cent of AI firms in Abu Dhabi focus on innovation, research and consultancy, according to the chamber.

From January to June 2025, 150 new AI companies were launched, driven by strategic investment, infrastructure and demand.

“Abu Dhabi’s artificial intelligence sector is rapidly evolving from early adoption to real-world transformation,” said Shamis Ali Khalfan Al Dhaheri, second VC and MD of the Abu Dhabi Chamber. “This surge is not only about numbers; it reflects a vibrant, diverse community of entrepreneurs, scientists and global leaders who recognise Abu Dhabi as a magnet for groundbreaking technology ventures.”

He added, “What sets this ecosystem apart is the strength of collaboration between government and business, global leaders and emerging innovators, and research and industry.”

The chamber has launched a 2025-2028 strategic roadmap focusing on ease of doing business, policy advocacy and ecosystem connectivity.

A dedicated Advocacy Working Group on Artificial Intelligence and Technology brings together sector leaders to shape Abu Dhabi’s AI future and enhance its competitive edge.

Few arenas combine raw human grit with technological brilliance quite like Formula 1. The roar of engines, precision manoeuvres, and split-second decisions have long been the hallmarks of this sport. But as the 2025 season charges ahead, a quiet shift is unfolding. Formula 1 (F1) isn’t just a battle of horsepower and courage anymore. It’s turning into a silent war waged with algorithms, neural networks, and cloud computing. With 1.5 terabytes of data generated per car per race weekend, the smartest team, not just the fastest, holds the edge. Welcome to F1’s new era: the code-driven arms race.

Let’s put this transformation into perspective. Each F1 car churns out 400GB of data per race, virtually enough to outpace the computing demands of an entire small business in a weekend. These raw metrics include telemetry, essentially the heart and pulse of the car’s performance, driver behavior analysis, tire pressure readings, and even real-time fuel consumption models.

The deeper truth? Cars today produce over one million data points per second during races. With teams crunching these numbers mid-race, decisions on vital elements like pit stops, tire changes, and fuel consumption are no longer instinctive, they’re informed by cutting-edge technology. When milliseconds dictate outcomes, precision is key, and AI has taken the wheel.

Gone are the days when a race engineer relied solely on experience and intuition to plan pit stops. Today’s AI systems evaluate over 150 parameters—from braking consistency and tire wear patterns to more nuanced metrics like driver stress responses to optimise each split-second maneuver.

It’s this obsessive command of data that allowed George Russell to squeeze out 97 per cent tire efficiency from medium compounds in 2023. In contrast, Lewis Hamilton achieved a still-impressive 94 per cent, highlighting how AI-guided insights are even differentiating performance among teammates.

Predictive analytics have transformed pit lane strategy. Through simulations and real-time learning, AI effectively answers questions before humans have thought to ask them: When is the exact moment to pit for maximum tire balance? How do fuel consumption rates shift as track temperatures climb? Which corner profile triggers driver fatigue? It’s a high-speed game of peering into the future playing out in milliseconds.

Racing simulations beyond the track

Step aside, traditional simulators; digital twins are in the game. These hyper-accurate simulations recreate cars and drivers down to their molecular behaviour, allowing teams to test countless strategies without ever setting a tire on the asphalt. The beauty of digital twins lies in their predictive value. By modelling whole car systems based on environmental inputs – humidity, track temperature, or even wind resistance – teams can anticipate performance shifts before they occur.

In many ways, digital twins embody F1’s transformation into not just a sport but a cutting-edge laboratory. The physical car becomes a manifestation of its virtual twin’s relentless experimentation. Could this mean the end of the once-revered gut instincts of drivers and engineers? Some followers of the “old F1” might argue so.

The silent race engineer

Edge computing has emerged as a game-changer in the pit lane. Teams can extract actionable insights with minimal latency by processing data locally during the race, without waiting for cloud-based solutions, teams can extract actionable insights with minimal latency.

Think of it this way: when you’re hurtling around a corner at 200 miles per hour, the difference between a half-second delay and instant feedback from the car can mean the difference between pole position and disaster.

As a key player in F1’s digital transformation, Globant’s Pitwall solution serves one crucial purpose: faster, more refined data delivery. Spectators can tap into real-time feeds of the analytics driving every lap, a digital experience as exciting as the race itself. Our collaboration with Formula 1 mirrors our work in other sports, such as FIFA and the LA Clippers. But our commitment to advancing AI-driven technologies within F1 sets us apart, highlighting how brands are becoming vital to the sport’s evolution.

The fan experience is also undergoing a transformation, thanks to augmented reality, predictive analytics, and interactive race streaming. Augmented reality overlays now provide intricate breakdowns of tire degradation and driver stress levels, all in real time.

Fans can witness firsthand the algorithms behind pit decisions, understanding in vivid detail why a driver switches from soft tires to mediums at a critical juncture.

F1 is no longer confined to the track, it’s flowing through the screens of millions worldwide.

Code versus courage: Losing the human element?

In this age of endless innovation, a simmering question remains: As Formula 1 becomes increasingly bespoke to AI, is the sport losing its human soul? Purists argue the sport’s essence lies not just in technology but in raw courage, the ability to take intuitive risks, to feel the vibrations of the car beneath you, to believe in a gut-driven moment that AI can’t quantify.

Yet, others counter this nostalgia by pointing to F1’s core appeal: competition. And if competition demands a smarter car rather than just a faster one, this evolution is simply logical. After all, making the driver-machine relationship stronger doesn’t dilute the sport, it enhances it.

F1 of the 2025 season is no longer just a race, it’s a high-speed chess match played between cloud infrastructure and edge computing, coded intuition and physical skill.

The millisecond decisions that once belonged solely to drivers and engineers now live within neural networks and predictive models. As the sun rises on this data-driven era, the car that wins isn’t just fast, it’s smart.

The CEO of New Markets at Globant

As Gulf states race to become digital hubs, two neighbours are carving very different approaches to data governance. Bahrain has invested in cloud and data-centre capacity and adopted a pragmatic, adequacy-style model for cross-border flows. Qatar is pursuing a centralised, sovereignty-first playbook—backed by a GDPR-style law, a state-led cloud framework and big infrastructure bets. Both trajectories create real commercial opportunities, but they also reshape compliance, cloud strategy and operational resilience for any firm doing business in the region.

That is the succinct read from Gareth Mills, partner at Charles Russell Speechlys, who has been advising clients on data, cloud outsourcing and cross-border transfers across the Gulf. “Qatar is positioning itself as a leader in digital sovereignty and regulatory readiness through a deliberate, top-down strategy,” he says. “This is primarily driven by the Qatar National Vision 2030 and the National Digital Agenda 2030 (NDA2030).”

Below is a practical, B2B guide to what Mills told us — why the two countries differ, where the enforcement risks lie, how banks and telcos should think about hosting and cloud, and what firms should do now to stay compliant and resilient.

Two contrasting strategies: centralised control vs. cloud enablement

Qatar is building a centrally governed digital stack. The state has layered a GDPR-style statute (the QPDPPL / Law No. 13 of 2016), guidance from the National Data Privacy Office, a Cloud Policy Framework that emphasises security (rather than blanket localisation) and investments in hyper-computing and national digital identity. Mills points to the coordinated, top-down nature of policy: the Ministry of Communications and Information Technology is playing a leading role, and the overall programme is designed to create both legal certainty and sovereign control over strategic digital assets.

Bahrain’s posture is different. As Mills explains, Bahrain “recognises the importance of data sovereignty but does not impose stringent data localisation requirements,” and it has sought to build local hosting capacity alongside a rules-based cross-border regime. The kingdom has actively attracted cloud and hyperscaler investment — AWS operates a regional data centre there — and private projects such as BEYON’s $700m “Digital City” further expand hosting and connectivity options. Bahrain’s Cloud Law also contains novel mechanisms — including “data embassy” arrangements that allow data stored in Bahrain to remain governed by the rules of another jurisdiction — boosting both flexibility and investor comfort.

The practical takeaway: Qatar is designing for sovereign control while enabling controlled openness; Bahrain is building cloud and data infrastructure and using an adequacy model to facilitate cross-border flows. Both are attractive, but your legal and technical strategy should match which regime applies to your licence and operations.

PDPL (Bahrain) vs QPDPPL (Qatar): the headline differences

Both laws share privacy fundamentals, but their operational shape is very different.

• Data localisation and transfers. As Mills notes, “Bahrain’s PDPL does not impose strict localisation mandates; there is no general requirement compelling organisations to store personal data within Bahraini territory.” Instead Bahrain relies on an adequacy list: only transfers to jurisdictions on that list (currently 83 countries) proceed without additional approval; other countries need PDPA authorisation or are managed via contractual safeguards. Qatar’s onshore statute, by contrast, “adopts a markedly permissive stance” — transfers may flow freely unless they would cause “serious damage” to data subjects, placing the burden on the exporter to assess and document risks.
• Enforcement approach. Bahrain’s PDPA has strong investigatory powers and a willingness to impose penalties, including fines and criminal sanctions for grave breaches. Qatar’s National Data Privacy Office historically took an education and guidance first approach, but Mills flags a change: since late 2024 regulators in Qatar have been taking a firmer enforcement stance, issuing binding decisions to correct material compliance gaps. The QFC (Qatar Financial Centre) meanwhile applies GDPR-style rules with clearer adequacy lists and contract-based transfer mechanisms.
• Practical consequence. In Bahrain expect prescriptive controls and active supervisory action; in Qatar expect a risk-assessment, documentation and DPIA-heavy model on the mainland, and a stricter, GDPR-aligned model inside the QFC. Organisations operating across both need a dual compliance track and consistent internal safeguards.

Licensing and registration: what businesses must do on day one

Mills highlights that obligations vary widely depending on the licence and the regulator:

• Qatar (dual system). Mainland entities governed by the QPDPPL must implement internal governance (a Personal Data Management System), maintain Records of Processing Activities and perform DPIAs for high-risk processing. There is no universal public register, but prior authorisation is required for processing “personal data of a special nature” (health, religion, children, criminal records). Entities in the QFC follow QFC DPR rules and the QFC DPO’s processes, which are closer to GDPR norms.
• Bahrain. Data controllers must register with the PDPA and notify processing activities. Data processors may also have registration duties depending on their role. Transfers to non-adequate jurisdictions require PDPA approval.

In practice: before you process any sensitive categories in either jurisdiction, map your licence (mainland vs free zone), compile RoPAs, embed DPIAs in project lifecycles and ensure you have documented approvals where required.

Sensitive personal data: sector implications

Both jurisdictions regard certain categories of data as especially high risk, but definitions and routes to lawful processing differ:

• Qatar (mainland and QFC divergence). Onshore Qatar defines “personal data of a special nature” to include health, religious beliefs, ethnic origin, criminal records and children; the QFC’s list broadens further to include political opinions and biometric data. Mills stresses the operational impact: “Healthcare: This sector faces the most stringent controls. Patient health information is sensitive under both regimes, mandating explicit consent and regulatory pre-approval … Telecoms: Operators must obtain explicit, opt-in consent for direct marketing …”.
• Bahrain. Processing sensitive data is generally prohibited without consent, except for enumerated exceptions (healthcare provision, public interest, legal claims, etc.). Financial services and telecoms must therefore build explicit consent mechanisms, robust security and carefully justified processing bases.

For regulated sectors such as healthcare, financial services and telcos, that means: pre-approval workflows (where required), enhanced technical protections, and rigorous consent and access controls.

Read: Data breach costs in Middle East drop 18% as AI adoption grows

Cross-border transfers: pick the right tool for the job

Mechanisms differ by jurisdiction and by licence:

• Qatar (mainland). There are no fixed standard contractual clauses mandated; instead exporters must document DPIAs and draft bespoke contractual protections. For particularly sensitive transfers, prior regulatory approval may be required.
• Qatar (QFC). Mirroring the EU model, the QFC recognises a list of “adequate” jurisdictions (EEA, UK, Canada, Japan, South Korea, Switzerland, Uruguay and California), and provides official SCCs and the option of BCRs.
• Bahrain. The PDPA’s adequacy list (83 countries) simplifies flows to those jurisdictions. Transfers to non-listed countries require PDPA authorisation and submission of contracts.

Mills’ practical rule: adopt a dual-track approach. Use DPIAs and tailored contract clauses for mainland Qatar flows; rely on QFC / PDPA adequacy mechanisms or SCCs/BCRs when operating under those regimes. Where feasible, align your internal policy to the stricter of the two frameworks — that simplifies governance and reduces legal friction.

Financial services and telecoms: local rules matter more than you think

Sector regulators impose additional constraints that often trump general data law:

• Qatar Central Bank (QCB). Retail banks and insurers face stringent localisation for customer data and tight cloud outsourcing rules. Material cloud outsourcing requires prior QCB approval and contractual terms that preserve supervisory access.
• QFC regulator (QFCRA). Wholesale firms may outsource to global cloud providers, subject to safeguards that preserve regulatory oversight and adequacy protections.
• Bahrain (CBB). The Central Bank of Bahrain mandates cloud and security standards for financial firms, which steers hosting choices toward providers with onshore capability.

Telcos in both jurisdictions must meet opt-in rules for marketing and special protections for children’s data. The upshot: cloud strategy must be sector-aware — a bank cannot rely on the same sourcing model as a non-regulated e-commerce operator.

Operational resilience and cloud contracts: the non-negotiables

Mills highlights the operational checklist regulators expect to see:

• ISO-level security (ISO 27001), encryption in transit and at rest, role-based access controls and multi-tenancy protections.
• Audit rights, SLAs with measurable recovery time objectives, and executable exit and portability plans to prevent vendor lock-in.
• Disaster recovery and business continuity plans that are demonstrable to the regulator.

For Qatari onshore entities, the regulator expects documented risk assessments for every cross-border transfer and active monitoring of third-party controls. For Bahrain, while operational resilience rules are still maturing, expectations are moving in the same direction. Practically: get your contracts right now (with audit and termination rights), and test failover plans periodically.

Law-enforcement access: prepare policies and playbooks

Both countries allow authorities to requisition data under national security and criminal law — often with wide discretion. As Mills summarises: “Qatar’s Cybercrime Law (2014), Telecom Law (2006), and Criminal Procedure Code empower national security and law enforcement agencies to access or intercept data, often without judicial oversight in security cases.” Bahrain likewise provides mechanisms for authorised inspectors to exercise law-enforcement powers.

Recommendation: establish a formal disclosure playbook — verification steps, proportionality checks, secure transfer controls and a rigorous logging regime. Limit disclosures to legal requirements, keep careful records, and train front-line staff to escalate any unusual requests.

Cross-border M&A, outsourcing and dispute readiness

Mills emphasises a commercial lens: data governance is now a deal and risk variable. Buyers will insist on strong RoPAs, evidence of DPIAs, encryption posture, and contractual remedies. Vendors must be able to show regulatory licences, approvals for sensitive processing, and tested incident response playbooks. For cross-border M&A and large outsourcing deals, the ability to demonstrate continuous compliance — not just a point-in-time audit — materially affects valuations.

Looking ahead

Mills believes that looking ahead, legal experts anticipate that GCC data governance frameworks will evolve rapidly, particularly in response to digital transformation and the rise of AI. Laws are likely to address ethical and privacy considerations around data usage and algorithm transparency, while cybersecurity regulations will tighten — potentially mandating minimum standards for firewalls, intrusion detection, encryption, and secure access controls, with sector-specific variations. For long-term resilience, businesses should invest in robust cybersecurity infrastructure, deploy technology solutions that support compliance, such as encryption and data management platforms, train employees regularly on data protection practices, and conduct periodic audits to identify gaps and vulnerabilities before regulators do.

Bottom line

Bahrain and Qatar are both actively building the region’s digital future but they do so from different starting points. Bahrain has doubled down on cloud capacity and an adequacy-style transfer model; Qatar is centralising governance, investing in sovereign digital infrastructure, and moving toward more assertive enforcement. For businesses that operate across the Gulf, that means designing compliance programs that are jurisdiction-aware, sector-sensitive and operationally hardened.

“Businesses should stay informed about these developments to ensure compliance,” Mills says. His practical advice is clear: treat data governance as a core part of commercial strategy, not a legal afterthought. Do that, and your cloud, outsourcing and cross-border plans will be ready for the next wave of Gulf digitalisation.

Definitions

Initial public offerings (IPOs) in the Middle East and North Africa (MENA) raised $2.5bn in Q2 2025, up 4 per cent from the previous quarter, driven largely by Saudi Arabia’s market activity, EY said in its latest MENA IPO Eye report.

Saudi Arabia accounted for 13 of the quarter’s 14 listings, raising a total of $1.9bn across sectors including transportation and healthcare.

Low-cost carrier flynas led proceeds, contributing 44 per cent of the quarter’s total, followed by Specialized Medical Company with $500m and United Carton Industries Company with $160m.

The UAE saw a single listing, Dubai Residential REIT, which raised $584m on the Dubai Financial Market. It became the largest real estate investment trust by market capitalisation in the Gulf Cooperation Council and the first pure-play residential leasing REIT in the region.

MENA region is a dynamic market for IPOs

“The second quarter of this year has reinforced the MENA region’s position as a resilient and dynamic IPO market,” said Brad Watson, MENA EY-Parthenon leader. “The diversity of sectors represented, along with milestone listings such as Dubai Residential REIT, highlights the depth of opportunities across the region.”

While 10 of the quarter’s IPOs closed below their offer price on debut, five recorded gains. EY noted issuers were increasingly strategic about timing, with 64.3 per cent of IPOs in Q2 being secondary listings, up from 35.7 pe cent in Q1.

“Saudi Arabia continues to set the pace for IPO activity in the MENA region, attracting strong interest across multiple sectors,” said Gregory Hughes, MENA EY-Parthenon IPO leader.

The Boursa Kuwait Premier Market Index led regional equity gains in Q2, up 17.2 per cent, while other markets posted mixed results.

The pipeline for Q2includes 14 expected listings, 10 from Saudi Arabia, with others planned in Egypt, Tunisia and Morocco.