Innovation and resilience remain key for tech industry in 2025 – Data and Analytics

In 2024, hybrid and multi-cloud computing, AI and automation, improving security and compliance, business resilience/efficiency and sustainability, were some of the key themes that drove tech investments.

While global markets have grappled with economic volatility, inflationary pressures, and geopolitical instability, technology spending within enterprises has remained relatively strong.

The accelerating need for digital transformation, remote work enablement, and enhanced customer experiences has fuelled investments in next-generation technologies, leading to continued growth in several key areas.

Moving into 2025, how will the industry evolve? What is the coming year like for enterprise tech adoption? How are tech decision makers planning their investments? Will it be more of the same from this year or different?

ITNews Asia speaks with practitioners to find out more about the issues and challenges companies face in tech adoption, how they are being addressed, and what will drive growth in the new year. Our respondents include: 

Tee Jyh Chong, Vice President, Sales and Services, APAC at Alcatel-Lucent Enterprise 

Joe Futty, Vice President of Product at Booking.com

Olaf Pietschner, Chief Revenue Officer, Asia Pacific Strategic Business Unit, Capgemini

Yasutaka Mizutani, APAC President, Colt Technology Services

Nick Dearden, Field Chief Technology Officer, Confluent

Peter Marrs, President, Asia Pacific, Japan & Greater China, Dell

Frank Bignone, VP & Global Director of Digital Transformation (DX) Division at FPT Software, FPT Corporation

Tan Ser Yean, CTO, IBM Singapore 

Daryush Ashjari, Chief Technology Officer, Asia-Pacific and Japan, Nutanix

Chris Chelliah, Senior Vice President, Technology and Customer Strategy, JPAC, Oracle

Matthew Oostveen, Vice President and Chief Technology Officer, Asia Pacific & Japan, Pure Storage

Yash Thakker, Director – Cloud Consulting, Searce 

iTNews Asia: What are the key technology challenges businesses are facing in 2024? What are the main barriers preventing businesses from fully adopting cloud native technologies or other next-gen technologies? 

Thakker (Searce): Migrating legacy systems to the cloud was difficult due to complexity, downtime risks, and data integrity concerns, while cyber threats demanded constant evolution in security measures to protect sensitive data.

Hybrid and multi-cloud environments further complicated adoption, as managing these ecosystems required expertise that many organisations lacked. Cost management remains a persistent issue, as businesses often encounter unpredictable pricing and mismanagement of resources, leading to overspending.

 

A widening skills gap in cloud-native technologies like Kubernetes, microservices, and serverless architectures slowed progress.

Pietschner (Capgemini): A major challenge is trusting the legacy systems, which makes it difficult to transition to modern cloud architectures. Shortage of skilled professionals in these technologies along with cultural resistance where employees may be reluctant to adopt new practices.

Cybersecurity remains a critical issue, with the shift to remote work increasing the attack surface. This risk is further heightened by generative AI, which not only boosts cyberattacks but also demands stronger defenses. 

Additionally, the integration of AI and automation can be complex, as organisations need to align new technologies with existing workflows while addressing ethical concerns like data privacy and algorithmic bias.

Yean (IBM): Many organisations face challenges such as integrating diverse data sources and cloud environments, unclear governance ownership, and a shortage of skilled talent.

An IBM study shows that while hybrid, multi-cloud strategies offer flexibility in data storage, only one third of technology and business leaders trust that AI solutions can be built and managed wherever their data is stored. This lack of trust hinders the widespread adoption of AI solutions. To overcome this, organisations must break down data silos and use integration tools to unify data sources.

Governance ownership is also a major issue. Only 18 percent of organisations have a dedicated AI and data governance role, and many struggle with navigating complex regulations and ensuring transparency across AI solutions. 

In ASEAN, many organisations also lack advanced AI and machine learning expertise, with only 17 percent having dedicated data science teams. 

Oostveen (Pure Storage): The key issue will be how to address the data challenges specifically, how to bridge data silos and protect data.

The other issue is whether they have the right people with the skills needed to turn these technologies into business value. We also believe budgets will tighten, and enterprises will prioritise projects with guaranteed ROI, meaning CIOs will be expected to do more with less.

Dearden (Confluent): Today, many enterprise organisations continue to leverage batch processing and reverse ETL (extract, transform, load) processes, and this data management paradigm is turning their systems into unmanageable “data systems spaghetti” with rising costs and diminishing value.  

Key challenges associated with this approach include:  

• Stale and inconsistent data, which can lead to insufficient data adoption and poor decision-making  
• Redundant processing and pipeline sprawl, which lead to increased operational costs and overheads 
• Long time-to-value for data engineering teams running business initiatives from complex batch-processing and redundant efforts  
• Sub-optimal innovation and growth with too many resources wasted on low-value activities like data cleaning and pipeline maintenance instead of building better customer experiences 

Ashjari (Nutanix): The landscape across Asia-Pacific and Japan is becoming more complex as organisations try to manage both legacy and modern applications across various IT environments, whether on-premises, at the edge, or in the cloud. This makes it harder for them to run applications and manage data anywhere.

Modern applications powered by AI need to be cloud-native and run on Kubernetes to enable faster development, fewer testing dependencies, and better portability. There’s an increasing need for a unified platform that supports Kubernetes and also manages legacy virtual machines.

The ongoing IT skills gap adds to these challenges, as organisations struggle to find talent with expertise in both modern and legacy systems, especially during day-2 operations of modern application environments.

Chelliah (Oracle): Many businesses struggle with infrastructure complexities as they operate in hybrid environments, with workloads spread across on-premises systems and multiple clouds. 

Currently, only 30 percent of enterprise workloads

have moved to the cloud, making it difficult for AI and machine learning to extract meaningful insights from fragmented data, limiting the benefits of automation and AI.

– Chris Chelliah, Senior Vice President, Technology and Customer Strategy, JPAC, Oracle

While the cloud’s agility is widely recognised, the high costs and time-consuming process of re-architecting legacy systems remain major obstacles. Migrating to the cloud often requires a complete rewrite of existing systems, creating both technical and financial challenges. 

Mizutani (Colt): Businesses faced challenges in integrating next-gen technologies like AI, SASE, and cloud-native applications with legacy systems and network security infrastructures.

Network security became the top priority for enterprise customers, especially as they ran AI or cloud-native applications with security-sensitive data. This created tension between maintaining existing operations and innovating for the future.

Data sovereignty and compliance requirements were also significant barriers, especially in regions with emerging regulations.

Bignone (FPT): Many organisations still rely on outdated, siloed systems. Migrating these can often disrupt business continuity. 

Cost management also remains a major hurdle, as the upfront costs of migration and ongoing multi-cloud management can be substantial. Transitioning to cloud-native architectures isn’t just a technical challenge – it requires a shift in company culture and workflows. 

At the same time, cultural resistance to change, fears of vendor lock-in, and concerns over regulatory compliance (especially in industries like healthcare and finance) make it harder to embrace cloud technologies fully. Many companies also struggle to quantify the ROI of cloud investments in the early stages, which can delay decision-making. 

Chelliah (Oracle): Organisations will also need to rethink their core systems to fit an AI-driven world. Legacy systems will evolve into “Systems of Intelligence and Interaction” that integrate AI to improve decision-making and automate tasks. 

This shift will streamline workflows, enhance transparency, and boost security, leading to more efficient and agile processes. By adjusting platform selection criteria, businesses can better capitalise on AI-driven innovation and efficiency.

Chong (ALE): ​​ Businesses now face challenges like integration complexity, the need for skilled workforce, and a changing regulatory landscape.

A key barrier to adopting cloud-native solutions is the complexity of integrating with legacy systems, along with the associated costs. With rising geopolitical tensions and evolving regulations across APAC, organisations must navigate complex compliance requirements.

Marrs (Dell): We are seeing strong momentum and fast progress towards AI readiness and deployment in the region, with more customers understanding the importance of data in this AI era. However, the fact remains that AI maturity in the regions and verticals differ market-to-market, given the highly diverse landscapes, rapidly evolving technology and resources available.

Customers are facing challenges in changing infrastructure as well as talent and service needs, with each organisation facing unique challenges to where they stand in the AI maturity scale. Many companies are also faced with organisational complexity, and are struggling to build the right organisation model that ensures that decisions made are top-down and anchored on strategic business objectives.

Futty (Booking.com): In the online travel industry, emerging technologies can help customers discover and plan their trips more effortlessly. Organisations often face issues like integrating AI with older systems, accessing reliable data, and building the right skills within their teams.

We have addressed such challenges through step-by-step processes, ensuring the use of AI or any new technology aligns with customer needs and business goals. They should be seen as a tool to solve real-world problems and deliver value and not just as a trend to follow.

iTNews Asia: Can traditional IT operating models (and legacy systems) align with new technologies and enable innovation. How critical is the need to change and modernise the IT infrastructure for 2025?

Oostveen (Pure Storage): While sticking with existing infrastructure may seem cost-effective in the short term, it can’t support growing needs. The massive data growth and other data-heavy applications highlights the limits of outdated storage systems, which can’t handle large volumes or advanced data formats. This hampers scalability and innovation.

Legacy systems also lack modern security features and energy efficiency. Without updates, they’re vulnerable to cyber threats. 

Chelliah (Oracle): Organisations will also need to rethink their core systems to fit an AI-driven world. Legacy systems will evolve into “Systems of Intelligence and Interaction” that integrate AI to improve decision-making and automate tasks. This shift will streamline workflows, enhance transparency, and boost security, leading to more efficient and agile processes. By adjusting platform selection criteria, businesses can better capitalise on AI-driven innovation and efficiency.

Chong (ALE): IT leaders need to move away from the mindset of “if it isn’t broken, don’t fix it.” Organisations can adopt hybrid models that combine traditional infrastructure with cloud-based solutions. 

Mizutani (Colt): While integration is possible, it’s often costly and complex. A hybrid or cloud-first strategy, along with solutions like SASE (Secure Access Service Edge) and SD-WAN, can help companies adapt quickly while improving network security.

Pietschner (Capgemini): Legacy systems, while deeply embedded in operations, are increasingly incompatible with new technologies like AI, cloud solutions, and advanced data analytics. These outdated systems, with siloed data and infrastructure, slow down real-time decision-making and hinder innovation.

Businesses must modernise their IT infrastructure by adopting flexible, modular architectures. Companies that don’t modernise risk falling behind and the need to innovate is no longer optional – it’s essential.

Bignone (FPT): Businesses can evolve rigid models by using key strategies. One approach is incremental evolution, where IT processes are gradually adapted, minimising disruption and allowing for smoother transitions. A hybrid model that combines legacy systems with cloud-native solutions ensures core functions are maintained while paving the way for future modernisation. 

For example, we provide solutions to help organisations modernise legacy systems like mainframes and COBOL codebases, enabling them to integrate new technologies with greater ease. Additionally, adopting APIs and microservices enhances the interoperability between legacy systems and modern applications, increasing flexibility and supporting global rollouts. 

Yean  (IBM): Modernising IT infrastructure is key for businesses to stay agile, competitive, and responsive to market changes. A major part of this modernisation is shifting from monolithic architecture to microservices.

While monolithic systems seem strong because of their rigidity, this very rigidity becomes a weakness when scaling. Even small changes may require dismantling and rebuilding the entire system, and these applications often need a complete overhaul to support new technologies.

– Tan Ser Yean, CTO, IBM Singapore 

Microservices, on the other hand, are designed for continuous growth and can easily adapt to new technologies. By breaking systems into smaller, independent services, each can be developed, deployed, and scaled separately. This flexibility allows businesses to automate their CI/CD processes, scale more easily, reduce downtime, and ultimately enhance the customer experience.

Dearden (Confluent): Companies that embrace real-time data streaming will lead in innovation and resilience, and the question is no longer if, but how quickly they can seize this opportunity. The key is to start small with high-ROI projects, keeping a long-term vision. This approach often creates momentum, accelerating further innovation and adoption.

Ashjari (Nutanix):

Traditional infrastructure struggles to manage complex Kubernetes environments for containerised apps, slowing down development, data protection, cost efficiency, and security compliance. Organisations that stick with legacy systems risk falling behind, missing out on opportunities with AI and next-gen technologies.

– Daryush Ashjari, Chief Technology Officer, Asia-Pacific and Japan, Nutanix

Modernising infrastructure is crucial for both operations and strategy, driving growth and innovation. A unified platform that supports both traditional and modern applications, and allows seamless movement between on-premises, cloud, and edge environments, will help businesses stay agile, flexible, and ready for growth in 2025.

Thakker (Searce): Traditional IT systems were built for specific tasks and struggle with the heavier workloads, modern apps, and large amounts of data we see today. Older systems often have outdated security and inflexible structures, which slow down innovation and make it hard to adapt to changing business needs.

Merging new tech with legacy systems is tricky and time-consuming and can lead to inefficiencies and missed opportunities.

To stay competitive in 2025, businesses must modernise their IT infrastructure adopting cloud-native solutions, automation, and advanced analytics. If companies don’t make this shift, they’ll fall behind in areas like agility, security, cost efficiency, and customer satisfaction. Businesses need a flexible infrastructure that supports innovations like cloud migration, DevOps, and AI/ML.

Futty (Booking.com): Aligning traditional IT models with new technologies is not easy – it requires a willingness to modernise, invest, and adapt. For instance, at Booking.com, we follow a ‘build-first’ mindset that allows us to develop and scale AI-powered solutions. Build first and learning along the way can help develop solutions that work at scale.

Modernisation is not just about keeping up but finding new ways to make it easier for everyone.

Marrs (Dell): We are evolving from static and reactive AI to a more dynamic, autonomous, interactive and profound set of tools that will allow us to go beyond what we have been able to do so far.

Agentic AI in particular, is a defining advancement in AI technology and will be a catalyst for significant progress in how we approach AI architecturally. The maturation of generative AI is giving rise to sophisticated AI agents capable of autonomous operation, natural language communication, and seamless collaboration with both humans and AI agents.

– Peter Marrs, President, Asia Pacific, Japan & Greater China, Dell

These specialised agents will possess a diverse range of skills, transforming industries and revolutionising workflows. The increasing complexity of these agent systems will necessitate the rapid evolution of technology stacks to support their diverse architectures.

iTNews Asia: What are the practical limitations of AI adoption that organisations should be aware of in 2025? Will companies look beyond experimenting in AI and start measuring their ROI?

Dearden (Confluent): Real-time data streaming is setting new benchmarks for innovation and resilience, making it clear that the question is no longer if companies will transform, but how quickly they can do so. Starting with high-ROI projects and maintaining a long-term vision will drive faster innovation and broader adoption.

Challenges to AI adoption include concerns over control, AI governance, and data complexity. Many worry about how AI/ML models will be controlled, especially as businesses balance open-source models (like Meta’s Llama) with proprietary ones (like OpenAI’s ChatGPT). 

– Nick Dearden, Field Chief Technology Officer, Confluent

Governance is another issue, as ensuring safe and responsible data access and sharing becomes more difficult with the growing number of global inputs. There is a need for stronger oversight to build trust in AI systems, though data governance frameworks remain fragmented. 

Additionally, the complexity of data within businesses poses a significant barrier to AI adoption, as data is often spread across multiple platforms and systems, making it harder to access accurate, timely information.

Ashjari (Nutanix): Infrastructure is a major barrier to effective AI adoption in Asia-Pacific and Japan. Many lack the robust systems needed for managing and processing data across different IT environments, and others are still deciding where to run their AI processes for the best results. Even once these decisions are made, moving applications between environments remains a challenge. 

As organisations move past AI experimentation, they need platforms that allow for efficient, scalable AI deployment. Hybrid multicloud solutions can solve these problems by offering a unified platform for running AI across various environments. This approach simplifies infrastructure, streamlines data protection, and frees organisations to focus on extracting business value from AI, rather than dealing with complex infrastructure issues.

Chong (ALE):

AI is moving from hype to reality, and companies are now focused on measuring its tangible benefits. However, realising these benefits depends largely on factors like data quality. Without structured, high-quality data, AI systems can’t provide reliable insights. 

– Tee Jyh Chong, Vice President, Sales and Services, APAC at Alcatel-Lucent Enterprise 

Additionally, businesses need to thoughtfully integrate AI into existing workflows, addressing concerns like employee resistance and understanding how AI can boost productivity. 

Yean (IBM): We anticipate that Asia-Pacific enterprises would move beyond AI experimentation to a hard-headed assessment of projected return on investment (ROI) in 2025. According to a recent IBM survey, more than half (54 percent) now expect AI to deliver longer-term benefits for their business in areas such as innovation or revenue generation. The game changer lies in developing cost-effective AI solutions, with the flexibility to use custom-built open-source models and ensuring seamless integrations between multiple vendors. 

Oostveen (Pure Storage): As AI matures, ROI will become even more important, pushing businesses to refine their strategies for tangible results.

Generic solutions like ChatGPT will see reduced use as concerns about output reliability grow. Instead, businesses will turn to more reliable models, like Retrieval-Augmented Generation (RAG), to ensure accurate, context-aware AI outputs.

This shift highlights the growing focus on transparency, ethics, and data integrity as organisations scale AI for real-world applications. Demand for RAG is expected to rise, especially in sectors like healthcare and financial services, where accurate, real-time data and context are crucial for informed decision-making.

Pietschner (Capgemini): As AI systems take on more decision-making roles, issues related to transparency, bias, and accountability must be carefully managed. Companies must recognise that AI is not a one-size-fits-all solution—it requires tailored implementation that considers both the technology’s capabilities and the specific needs of the business. Moving beyond experimentation, organisations will need to focus on the scalability of their AI efforts and, most importantly, measure the tangible impact on their bottom line.​ 

Chelliah (Oracle): Many organisations will expect quick results and may scale back their investments too soon if they don’t see immediate impact. Patience and clear alignment between AI initiatives and business goals will be crucial. Leaders who focus on long-term innovation and integration will unlock the full potential of AI.

The growing number of AI solutions will also make decision-making more complex. With so many options, businesses must prioritise quality over quantity to avoid overwhelming testing, security, and cost issues. Success will come from carefully selecting AI tools that offer the most value.

Mizutani (Colt): As AI use cases mature, it is becoming a central driver in digital transformation for businesses. We expect more companies to adopt multicloud solutions with high-speed, secure, and flexible networks, with edge AI emerging as a key use case due to data sovereignty and low-latency needs.

The regulatory landscape is also crucial, with countries like Japan, Singapore, India, and the EU introducing policies and frameworks to guide AI adoption. Businesses will rely on these regulations to navigate AI’s societal and business impacts.

– Yasutaka Mizutani, APAC President, Colt Technology Service

Many companies are already seeing strong returns from AI, including productivity gains, cost savings, better customer satisfaction, and fewer errors. While measuring ROI is important, it’s also essential to consider the risks of AI adoption to ensure its ethical use and long-term success. 

Thakker (Searce): In 2025, data quality and availability will be a major challenge for AI, as models require large, clean datasets that are hard to obtain and maintain. Other issues include managing and updating AI models, dealing with latency, and ensuring that models are interpretable. 

Many organisations also lack the AI expertise needed to implement and manage complex models, which can lead to delays or poor results. Integrating AI with legacy systems is another hurdle, as older infrastructure may need costly upgrades.

To succeed, companies should focus on high-impact, low-risk AI use cases, align efforts with clear business goals, and measure performance before and after AI implementation to justify further investments and embed AI into their core operations.

– Yash Thakker, Director – Cloud Consulting, Searce 

Bignone (FPT): Expanding AI from pilot projects to full-scale enterprise operations requires significant investments in infrastructure and skilled talent, as well as extensive change management efforts to align teams, processes, and technologies.

Ultimately, overcoming these hurdles requires selecting the right partners with expertise in AI, data management, and regulatory compliance. 

As businesses move towards 2025, the focus will shift from experimental AI projects to pragmatic, results-driven initiatives, demanding proof of AI’s value through clear metrics like operational efficiency, customer satisfaction, and revenue growth.

– Frank Bignone, VP & Global Director of Digital Transformation (DX) Division at FPT Software, FPT Corporation

A phased implementation approach will help manage resources, minimise risks, and ensure a solid return on investment.

Marrs (Dell): 2025 will be the year we see enterprise AI truly scale and get into full production. Organisations are already increasingly prioritising tangible ROI and business value from their AI initiatives, often establishing dedicated AI committees led by Chief AI Officers or CIOs. This pragmatic approach is reflected in the rising success rates of GenAI pilot projects.

However, enterprises must be aware of the talent and resource gaps in the industry, and recognise the urgent need for AI skills development. The future of AI hinges on collaboration – between humans and AI, and between organisations and their technology partners.

Futty (Booking.com):

Using AI responsibly and embedding fairness, transparency, and accountability is difficult yet essential.

– Joe Futty, Vice President of Product, Booking.com

For instance, at Booking.com, our AI-powered features like AI Trip Planner or Smart Q&A – is carefully tested to meet high standards of quality and trust. Such efforts can be supported by extensive customer research.

iTNews Asia: What steps should businesses take to ensure better data governance and protect sensitive information in 2025?

Bignone (FPT): Organisations with robust data governance frameworks will be better equipped to navigate complex regulations and minimise risks.

Key steps include adopting federated governance models, which allow departments to manage their own data within a unified framework; implementing privacy-by-design practices to prioritise privacy at every stage of data handling; leveraging automation for compliance tasks like data discovery and monitoring to reduce errors and speed up compliance; and fostering cross-functional collaboration to make data security everyone’s responsibility. 

Regular audits, real-time monitoring, and ongoing training should also be prioritised to maintain data governance and safeguard sensitive information.

Ashjari (Nutanix): As more organisations deploy applications across different IT environments, they need a consistent approach to data management and governance, whether their applications run on-premises, at the edge, or in the cloud. 

A unified hybrid multicloud platform, combined with strong data protection and cybersecurity measures, simplifies data management and streamlines backup and recovery across different infrastructures. As organisations adopt AI, they need scalable backup and recovery solutions that can handle increasing data volumes without sacrificing performance.

Additionally, a comprehensive approach to backup and data recovery is essential. This includes robust encryption, access controls, and anomaly detection to protect data integrity and availability. With cybercriminals targeting data more than ever, organisations must prioritise rapid recovery from data losses or breaches, while ensuring high availability and compliance with data privacy and protection regulations.

Oostveen (Pure Storage):

A successful cybersecurity strategy requires both people and technology to work together. Even with advanced systems in place, data breaches still happen due to human error.

– Matthew Oostveen, Vice President and Chief Technology Officer, Asia Pacific & Japan, Pure Storage

To address this, organisations should take a two-pronged approach. First, from a people perspective, regular training is essential at all employee levels. This includes educating staff on data hygiene, how to spot phishing attempts, and how to handle sensitive information. Security and IT teams should also be trained to respond to increasingly sophisticated cyber threats.

Second, from a technology standpoint, businesses should implement a comprehensive defense strategy that includes strong data protection and rapid recovery capabilities, ensuring operations can resume quickly after an incident.

Dearden (Confluent): Fostering a culture of security and compliance in organisations requires a multi-faceted approach. First, improving data quality and integrity is key. This also allows for better application of security measures, such as hiding, redacting, or encrypting sensitive data. As organisations create more governed, trustworthy data products, these become valuable, reusable assets that can drive better business outcomes.

Secondly, enhancing employee data literacy is essential. This not only boosts employee productivity but also ensures greater awareness and compliance with security and privacy safeguards.

iTNews Asia: While AI can streamline operations, it also increases vulnerabilities. How should you prepare your cyber security strategy to protect against AI-led cyber attacks?

Pietschner (Capgemini): Our research shows that nearly all organisations have experienced breaches linked to AI, highlighting the urgent need for stronger defenses.

To counter these threats, businesses must update their cybersecurity strategies. A multi-layered defense, powered by AI and machine learning, is crucial for detecting and responding to threats in real-time. Automation can help improve threat intelligence and response times, easing the load on human security teams.

As businesses adopt more autonomous AI systems, they must focus not only on technical protections but also on building trust in these systems. Additionally, preparing for the rise of quantum computing by investing in post-quantum encryption will be key to safeguarding data in the future.

– Olaf Pietschner, Chief Revenue Officer, Asia Pacific Strategic Business Unit, Capgemini

Bignone (FPT): AI-driven cyberattacks, such as deepfake phishing, automated exploits, and AI-powered malware, require businesses to adopt a proactive, adaptive, and multi-layered cybersecurity strategy. As these threats evolve, organisations must combine advanced technology, strong governance, and human vigilance.

Key strategies include leveraging AI for defense to detect anomalies and automate threat responses, adopting zero-trust models to continuously verify access and reduce internal and external threats, and securing AI systems against adversarial attacks by implementing robust security protocols like model validation and data integrity checks. 

Additionally, training teams to recognise AI-enhanced threats is crucial. FPT, for example, integrates AI into operations and provides employee training while collaborating with research institutes to foster AI talent in Vietnam. 

Futty (Booking.com): AI-driven cyberattacks are a growing concern as attackers become more sophisticated. To tackle this, businesses need to use AI as a powerful defense tool, capable of identifying and stopping threats quickly and effectively.

At Booking.com, we had intercepted 1.5 million phishing-related bookings and blocked 85 million fraudulent reservations. These numbers show the scale of the challenge. While no system is perfect, companies must continuously improve the processes and technology to stay ahead of cyber threats.