Lebanon and Israel said on Wednesday that they had agreed to the implementation of a ceasefire following negotiations in Washington aimed at ending the conflict that flared up alongside the war in Iran.

The ceasefire is contingent on a complete cessation of fire from Hezbollah and the evacuation of all of its operatives from the South Litani Sector, according to a joint statement released with the United States.

“The two sides agreed with the guidance of the United States to swiftly advance the creation of pilot zones in which the Lebanese Armed Forces will take exclusive control of the territory to the exclusion of all non-state actors,” they said.

Israel and Lebanon had previously agreed to a cessation of hostilities in April that was then extended in May, but violence has continued.

Israeli drone strikes killed at least six people in southern Lebanon and targeted a car south of Beirut ​on Wednesday, Lebanese security sources said. Israel said it intercepted a hostile aircraft likely fired by Hezbollah.

A US-mediated agreement announced on Monday had led Israel to step back from attacking the Hezbollah-controlled southern suburbs of Beirut, and the Iran-backed group to halt cross-border strikes.

Israel invaded Lebanon in March in pursuit of Hezbollah which fired across the border in support of Tehran.

Iran has said it will not agree to a deal to end the conflict with the United States and Israel launched in late February unless a ceasefire also covers Lebanon.

Lebanon and Israel on Wednesday agreed to further direct negotiations to build confidence and resolve other outstanding issues.

The pair will reconvene to hold political and security-related talks during the week of June 22 with a view toward a comprehensive agreement, according to the statement.

The rise of “vibe coding”, AI tools that can spin up working applications from a plain-language description, has reopened a familiar debate about who gets to build software, and how much of it can be automated. For Hyther Nizam, CEO of Zoho MEA, the shift is real but often misread. He sees AI generation as a genuine breakthrough in getting from an idea to a working prototype in minutes, putting that power in the hands of finance analysts and operations managers rather than developers alone. The harder problem, he argues, is everything that comes after the first build: maintaining interdependent workflows, approval chains, user roles, and compliance requirements as business rules change and applications scale.

In this conversation, Nizam draws the line between fast AI-assisted generation and structured low-code platforms like Zoho Creator, explains why security, auditability, and data residency have become procurement requirements rather than nice-to-haves, particularly for regulated industries in the UAE, and makes the case that traditional coding, low-code, and vibe coding are heading toward a division of labour rather than a winner-takes-all outcome. The thread running through it all: keep a human in the loop.

There’s growing talk about “vibe coding” and AI tools that can generate applications with minimal human input. Do you see this as a real shift in software development, or more of an early-stage experiment?

Vibe coding is the most significant change in who can build software in the last decade. The ability for an operations manager or a finance analyst to describe a process and have a working interface generated in minutes is not a prototype feature; it is a structural shift in how organisations will approach internal tooling.

What it doesn’t solve is what happens after the first build, particularly for serious applications that organisations actually run on. Enterprise software is a system of interdependent workflows, approval chains, user roles, and compliance requirements. Maintaining that over time, as business rules change and the application scales, is where vibe-coded outputs start to show their limits. For instance, the trail of changes made, components altered, code written and rewritten, and technology used are all outside the absolute control of a developer, resulting in a steep governance risk.

Where do you draw the line between AI-assisted development and structured low-code platforms like Zoho’s? What does each do better in practice?

AI-assisted development is good at application generation. Users can describe intent and get a prototype fast. Low-code platforms, on the other hand, combine ease of development with the layer underneath — the underlying data model, workflow logic, user permissions, security, compliance, integrations, and more — that is most critical for enterprise adoption. Low-code also lends itself to code optimisation in a way that free-form AI generation doesn’t, because the underlying structure is defined and inspectable rather than generated from scratch each time.

The distinction shows up most clearly over time. AI tools help you go from zero to something working quickly. Low-code platforms help ensure that something survives contact with a real organisation, bringing version control, role-based access, audit trails, and the ability for an operations lead to modify their own workflow without raising a development ticket. The two are increasingly complementary rather than competing.

One argument is that AI could eventually replace traditional and even low-code development entirely. What is your view on that trajectory over the next five to ten years?

Writing code was a real bottleneck, and AI has genuinely reduced it. What it hasn’t resolved is the broader challenge: whether an AI-generated application can be maintained reliably, scaled as the organisation grows, and kept relevant as business requirements evolve. Those are open questions, and until they’re answered confidently, human control remains essential in the development process.

At the pace at which things are evolving, predicting the situation over the next five to ten years is a dart thrown in the dark. That said, we expect AI to become a standard capability inside all modes of development tools, including low-code platforms. What it will not replace is domain expertise. The person who best understands how a procurement approval process should work is not a developer but the procurement manager. LLM-powered tools can act as an efficient support system, but the direction and action should be owned by the human in the loop.

Zoho has long invested in low-code as part of its broader software ecosystem. How is AI changing what low-code means inside your own product roadmap?

We’re introducing a new unified development environment that embeds AI across the entire software development lifecycle, from requirements and build through to testing and in-app agent creation. The intent is not to replace the structured low-code environment but to make it significantly faster to work within it, while preserving the human-in-the-loop approach that ensures the integrity of what gets built. AI handles the generation and suggestion; the developer retains control over what gets committed.

The foundation remains the same: Zoho’s low-code platform sits as the process layer across the broader Zoho ecosystem, connecting your applications to the same live data environment as your CRM, finance tools, and HR systems. AI capabilities are being layered on top of that structure, not in place of it.

From an enterprise perspective, how do concerns around security, governance, and scalability shape the case for low-code versus fully AI-generated applications?

For enterprise buyers, security, auditability, and scalability aren’t differentiators. They’re table stakes. A useful rule of thumb is this: if security, compliance, and clear accountability need to be in place, a governed platform is the right foundation. With fully AI-generated applications, that burden falls back on the organisation — covering security review, role testing, and ongoing maintenance as business rules change — each requiring developer time that many enterprises are already short of.

A structured low-code platform shifts that burden into the platform itself. Version control, role-based access, audit trails, and data residency controls are available by default in Zoho Creator, not things a user needs to configure from scratch. The more relevant scalability question today is whether your ops team, finance team, and regional offices can all build and adapt their own workflows without creating a bottleneck at the centre. That is where a purpose-built platform has a structural advantage that AI generation alone doesn’t address.

Are you seeing enterprises in the UAE and wider region actually move from traditional development to low-code, or is adoption still limited to specific use cases?

Adoption in the region has moved well beyond specific use cases. What’s changed is the ambition of the implementations. We’re now seeing enterprises use Zoho’s low-code platform as the backbone of their operational infrastructure, covering field inspection management, multi-entity compliance tracking, partner portal development, and custom reporting platforms that pull live data from across their Zoho environment.

The UAE specifically has seen accelerated adoption driven by data sovereignty requirements and smart government initiatives. Zoho operates its own data centres in Abu Dhabi and Dubai, which means enterprises build and run applications with their data remaining in-country. For regulated industries in financial services, healthcare, and government, that is not a nice-to-have; it is a procurement requirement. Enterprises already on Zoho CRM, Books, or Desk find that Creator applications running within that same sovereign infrastructure eliminate significant integration overhead and dramatically shorten the path to production.

Where does low-code still struggle today, especially when compared to newer AI-native development tools?

Low-code has a learning curve, and in terms of raw speed of the initial build, AI generation is faster for simple use cases. That is a fair observation. But the comparison changes significantly once you factor in maintainability, because the structured environment makes ongoing changes inspectable and predictable in a way that AI-generated output typically isn’t.

Most low-code platforms follow a human-in-the-loop approach by design, and that exists to protect the integrity of the application being built, not to slow it down. For applications that an organisation genuinely runs on — connected to live financial or operational data and expected to evolve — that integrity is non-negotiable. The first build is rarely the expensive part; keeping it working and adapting it over two or three years is where the real cost sits.

If we look ahead, what does the “ideal” development stack look like in a world where AI, low-code, and traditional coding all coexist?

There is definitely a space for all three, and the organisations ahead in the AI adoption curve are already treating it as a division of labour rather than a competition. The use case and developer persona will determine the choice of approach, and in most scenarios, a combination of all three will be preferred over any single mode, with AI embedded natively across each.

For now, AI-native vibe-coding platforms will serve non-technical business users who need rapid prototyping and deployment for simple internal workflows, lightweight automations, and proofs of concept. Traditional and low-code platforms will remain the choice of professional and techno-functional developers building org-wide solutions where compliance, security, scalability, and maintainability are non-negotiable. What low-code adds is ownership at the process layer, where business workflows can be built, maintained, and adapted directly by the people closest to the work, without routing every change through an engineering team. This helps organisations strike a balance between faster development and enterprise governance requirements.

The practical risk is simpler than it sounds. When business teams can’t get what they need, they build outside official channels. When enterprises over-rely on AI generation without a governed layer, they build fast and maintain slowly. When one tool gets mandated for everything, it fits nothing well. The right stack isn’t a single answer; it’s the discipline to match the right tool to the right problem — and the teams that get this right are the ones that build durable operational capability rather than cycling through rebuilds every 18 months.

Read: Building trust in AI: The UAE’s journey to a digital cognitive future

Al Ain Farms Group (AAFG), one of the UAE’s largest food and beverage manufacturers, has committed to reducing added sugars by between 10 per cent and 20per cent across its brands as part of Abu Dhabi’s Healthy Living Strategy, the company said on Monday.

The initiative covers brands including Al Ain Farms and Marmum Dairy and aims to improve the nutritional profile of widely consumed food and beverages while maintaining product quality, taste and consumer trust.

The reformulation programme is one of 28 strategic initiatives under Abu Dhabi’s Healthy Living Strategy, which seeks to make healthier choices more accessible through system-wide interventions.

The commitment was formalised in collaboration with Healthy Living and the Abu Dhabi Quality and Conformity Council (QCC), marking a groupwide effort to enhance nutrition across AAFG’s dairy and beverage portfolio.

Dr Ahmed AlKhazraji, executive director of Healthy Living, said improving the nutritional quality of everyday food products was one of the most effective ways to support healthier lifestyles at scale.

“We want to ensure that healthy, nutritious food is within everyone’s reach – because eating well should be easy for all,” he said, adding that the initiative would help make healthier options more widely available.

Engineer Abdulla Hassan Al Muaini, executive director of the Central Testing Laboratory at QCC, welcomed the commitment, saying it aligned with efforts to help consumers make informed choices without compromising on quality or taste.

Hassan Safi, Group CEO of AAFG, said the company was combining product innovation with reformulation across its portfolio to make healthier choices more accessible while maintaining taste and quality.

Read: The making of a ‘National Champion’: How Al Ain Farms Group is nurturing the UAE’s food future

AAFG launched a new healthy product range in April 2026 and plans to reformulate products, including flavoured milk, yoghurt and laban sold under the Al Ain Farms and Marmum Dairy brands.

The updated products are expected to be available before the back-to-school season in September 2026.

The company said the initiative demonstrates how local food manufacturers can support government efforts to improve nutrition and consumer wellbeing, particularly among children and young people.

Kuwait could restore nearly 70 per cent of its oil production within six to eight weeks after the Strait of Hormuz reopens, Kuwait Petroleum Company’s managing director for international marketing, Shaikh Khaled Ahmad Al-Sabah, said on Wednesday.

The remaining 30 per cent would take about another month, he told the S&P Global Energy Middle East Petroleum and Gas Conference.

Kuwait’s timeline for its production recovery is shorter than some forecasts for a full reopening of transits through the Strait of Hormuz, which Iran has effectively closed since U.S.-Israeli attacks in late February.

On Tuesday, ADNOC’s executive vice president for sales and trading Philippe Khoury said full transits through the strait could take until mid-2027 to recover to pre-war levels. The International Energy Agency’s head of oil, Toril Bosoni, said a recovery could take six to eight months in the best-case scenario from now if an agreement was reached.

Separately, Al-Sabah said KPC could restore its refinery output to normal levels in around two to three weeks. KPC has about 1.4 million barrels per day of refining capacity, he said.

Vitol Bahrain’s head of research, Bader Nooruddin, forecast on Wednesday that Gulf refineries could ramp up to about 90–95 per cent of capacity within 40 to 60 days.

Middle East refiners are already planning for a future after the current supply crisis.

Al-Sabah said Kuwait is in talks with “friendly countries” on potential pipeline projects.

“A lot of people thought, why build a pipeline without using it? Now shows the use of a pipeline,” he said, adding the crisis had also highlighted Kuwait’s need for larger storage capacity.

Austrian oil firm OMV echoed the comments, with general manager Mikael Berthod telling the conference that Middle Eastern refiners must become more commercially agile and invest in pipelines and storage over the next two to three years.

They will also need stronger partnerships to handle future supply shocks, he added. OMV has investments in the Middle East.

In the near term, ADNOC expects a spike in oil demand to rebuild inventories, followed by a steady recovery as prices normalise, senior vice president of business transformation Fatema Bin Saleem Al Teneiji said.

Dubai’s Knowledge and Human Development Authority (KHDA) will resume quality assurance visits to private schools across the emirate from the 2026-27 academic year, marking a significant step in the emirate’s efforts to enhance educational standards, improve student outcomes and strengthen parental confidence in the private education sector.

The initiative forms part of Dubai’s Education 33 (E33) Strategy, which seeks to elevate the quality of education while placing students at the centre of the learning experience, a WAM report said.

The strategy also supports wider objectives outlined in the Dubai Plan 2033, Dubai Economic Agenda (D33) and Dubai Social Agenda, all of which position education as a key pillar of the emirate’s long-term development.

Focus on school improvement

The Education Quality Assurance and Compliance Agency, a division of KHDA, said the renewed programme will reinforce oversight of private education providers while enhancing quality assurance mechanisms designed to support continuous school improvement.

Read-Dubai halts private school fee hikes for 2026-27 academic year

Over the past two academic years, KHDA maintained oversight of educational quality through targeted visits to newly established schools and institutions completing their first three years of operation. The authority also relied on school self-evaluation reports, data analysis and student performance assessments to monitor standards across the sector.

Beginning in the 2026-27 academic year, quality assurance activities will place greater emphasis on evaluating the impact of improvement measures implemented by schools during the past two years. The programme will also provide targeted guidance to help schools build on progress and address areas requiring further development.

Two-tier inspection model introduced

Under the revised framework, eligible private schools will be assigned one of two forms of quality assurance visits.

The first category will involve a comprehensive inspection conducted by specialist teams using the UAE School Inspection Framework. Schools undergoing a full inspection will receive a detailed report that includes an overall performance rating. Schools completing their third year of operation in Dubai will automatically be subject to a full inspection.

The second category will consist of shorter monitoring visits focused on specific areas identified through performance data and analysis. These visits will result in concise reports highlighting strengths and recommendations for improvement, although no new overall rating will be issued.

KHDA said schools selected for either type of visit will receive no more than 24 hours’ notice. The approach is intended to ensure inspections accurately reflect the day-to-day reality of school operations, teaching quality, learning environments and student wellbeing.

Building confidence in Dubai’s education sector

Fatma Belrehif, Chief Executive Officer of the Education Quality Assurance and Compliance Agency, said the renewed framework is designed to strengthen confidence among parents while supporting the broader goals of the Education 33 Strategy.

She said the differentiated inspection model recognises that schools are at varying stages of development and enables inspectors to focus on areas most closely linked to student achievement and educational outcomes.

The quality assurance programme will continue to operate under the UAE School Inspection Framework, which was introduced during the 2015-16 academic year. KHDA said it will use a moderated, data-driven methodology to determine the most appropriate type of visit for each school, drawing on performance metrics, self-evaluation processes and other key indicators.

The authority added that the initiative supports Dubai’s ambition to strengthen its position as a global hub for high-quality education by advancing educational outcomes, promoting equity and driving continuous improvement across the private school sector.