Iranian Attacks on GCC Data Centers Put Tech Infrastructure Investment Risks into Sharp Focus

by Ismail Patel, posted in Big Data
A person with a beard speaking into a microphone at a conference, with an audience in the background.
I. Patel

Summary Bullets:

• Military strikes by Iran damaged AWS data centers in the UAE and Bahrain, highlighting vulnerability of the region’s data centers.

• Foreign investors, local tech decision makers, and insurance firms are now revisiting the calculus of risk in the data center boom.

Earlier this week, AWS data centers – two Availability Zones (AZ) in the UAE and another in Bahrain, all part of the AWS ME-Central-1 Region – suffered physical damage in escalatory strikes by Iran. The attacks disrupted 73 core services in the Bahrain AZ; at the time of writing, as per information shared by the AWS Health Dashboard, AWS stated that only 33 services were resolved. In the UAE, of the 112 impacted AWS services, only three have been resolved while others still face severe disruption or degradation. AWS strongly advised customers to enact disaster recovery plans and migrate workloads to unaffected AWS Regions in Europe, APAC, and the US.

The immediate impact of the tech outages has been on regional banking platforms (e.g., ADCB, Emirates NBD), consumer apps (e.g., Careem), enterprise software platforms (e.g., Snowflake), payment apps (e.g., Alaa, Hubpay), and government services were affected. More importantly, the incidents have undermined confidence in the long-term viability of large-scale US-led physical infrastructure investments. In some ways, data centers are the new oilfields, in that they are targets for hostile actors.

The recent events come on the back of the Gulf states – particularly the UAE and Saudi Arabia – courting large-scale AI data center investments from American vendors. Governments have emphasized controlled tech import regimes, and supply-chain integrity that mitigated the risk of proprietary American tech being leaked to China or Russia. Last year, the GCC nations made vast commitments to ambitious AI and data center projects exceeding $100 billion. Cybersecurity has been a hot topic over the past few years. But what was relegated to the background here – until now – was the physical integrity of GCC’s data center infrastructure from kinetic threats, which have suddenly sprung up as a result of the latest tensions. Whatever the short-, medium-, and long-term outcomes, this dynamic adds a new dimension of concern to the tech leadership ambitions of the GCC nations.

For cloud data centers, there will be an architectural and operational rethink by the American hyperscalers. Reliance on multiple AZs within the same region is evidently insufficient. Resilient systems will demand geo-agnostic, intercontinental diversity, with cloud providers being able to demonstrate broad critical workload distribution to defend against region-level attacks. The impact of current events means operational and business continuity planning will shift from best practice to necessity, underpinned by redress for reputational damage or economic failure. Still, sectors like finance and healthcare and single-region deployment that require complete sovereignty will be exposed to greater risk and regulators will be thinking hard on how best to mitigate vulnerabilities.

For the promised mega AI data centers that are at the beginning of their build outs in Saudi Arabia, UAE, and Qatar, investors will most probably be looking carefully at the commitment to the scale of investments that were promised in 2025. The region still possesses a lower tax burden, lower energy costs, vast greenfield opportunities, and governments willing to make things happen. However, it is expected that any new data centers will require physical hardening through blast shielding or by potentially moving underground. They will also drive up OPEX with higher insurance costs. All of this has the potential to shrink the ambitious scale of the build outs. To assist, regional governments are likely to establish formal protection procedures for digital infrastructure hosted in their borders in recognition of AI and cloud data centers as critical national assets.

RingCentral Strengthens Two Contact Center Offers, Courtesy of OpenAI

by Gregg Willsky, posted in Big Data
A smiling man with glasses, wearing a light-colored shirt, with a warm expression.
G. Willsky

ummary Bullets:

• The infusion of OpenAI Frontier models into AIR and AVA marks a meaningful step for RingCentral in the contact center space.

• The announcement suffers some blemishes despite the positives. RingCentral should deliver functionality faster and provide adequate supporting detail.

RingCentral is infusing OpenAI Frontier models including GPT-5.2 into two voice-centric contact center offers: RingCentral AI Receptionist (AIR) and RingCentral AI Virtual Assistant (AVA). OpenAI Frontier is an enterprise platform introduced in February 2026 that enables organizations to build, deploy, and manage AI agents that access internal systems, share context, and execute inter-departmental workflows.

AIR answers calls, handles tasks such as scheduling appointments, conducts follow ups through human-like conversations, and routes inquiries. AVA is the company’s new personal AI assistant currently available to a limited number of customers. AVA picks up where AIR leaves off. AIR acts as a ‘front door’ executing what it can and handing off inquiries through AVA as necessary. AVA channels those inquiries to employees along with full context, customer insights, and a choice of next-best actions. In doing so, AVA connects voice, video, and messaging data across the RingCentral platform, capturing notes and summarizing calls, searching across conversations, automating follow-up tasks and the like.

The infusion of OpenAI Frontier models into AIR and AVA marks a meaningful step for RingCentral in the contact center space. In the last few years contact centers have profoundly transformed, steadily yielding to the broader concept of ‘customer experience’. Contact centers are converting from featuring live agents to also including AI agents, from reactive to proactive, from transaction-oriented to relationship-oriented, and from generic to deeply personalized. RingCentral and its rivals continue to implement capabilities to help their customers make the transition.

The RingCentral announcement reflects an additional way contact centers are transforming. The AI agents that have become a fixture of contact centers are assuming increased responsibility for resolving customer issues. They are pivoting from a tool that can only answer basic questions to one that can also execute tasks as well as engage subject matter experts outside the contact center when needed.

The announcement suffers some blemishes despite the positives. While the new capabilities are beneficial and in keeping with contact center trends, they are already collectively reflected by rivals such as Cisco, Zoom, and Mitel. In addition, not specifying a general availability date for AVA equates to a missed opportunity to build some market anticipation. Lastly, omitting pricing details only generates confusion. If RingCentral can deliver functionality faster and provide adequate supporting detail it will go a long way towards making introductions more impactful.

Though it should be noted that RingCentral and competitors are opening a Pandora’s box of sorts. While deploying AI agents that are more flexible than their ancestors can close out customer issues more efficiently and effectively, it also raises the possibility of these agents wreaking havoc; for example, booking a customer on the wrong flight or cancelling an order instead of rescheduling the delivery date. Vendors need to ensure that appropriate safeguards are in place.

AMD and Meta’s Deal is One of the Most Significant AI Infrastructure Partnerships This Year, but Will It Deliver?

by Beatriz Valle, posted in Big Data
A professional woman with brown hair, wearing a black blazer and a teal top, against a light blue background.
B. Valle

Summary Bullets:

• On February 24, 2026, AMD and Meta announced a partnership to deploy AMD Helios racks, optimized for Meta’s workloads.

• The shipments, starting in H2 2026, will cover successive generations of silicon over several years and be equivalent to 6 gigawatts of power.

AMD and Meta are deepening their collaboration to align their GPU and CPU silicon, systems, and software roadmaps with Helios rack clusters running on ROCm software. As part of the agreement, AMD is also giving Meta warrants that could convert into a 10% stake in the company. Meta can only cash the warrants if it buys all the agreed chips, and AMD’s share price triples. For AMD it’s a massive validation of its AI computing roadmap and for the wider industry it has broad implications as it can mean lessening overreliance on a single supplier, potentially accelerating innovation. It gives Meta greater bargaining power as it gains pricing leverage and potentially reduces the risk of supply bottlenecks. In other words, Meta avoids being completely locked into NVIDIA’s CUDA ecosystem.

Comparisons have been drawn between a similar agreement signed by AMD and OpenAI in October last year. However, this is different. The terms of the deal may be similar; the scale is not. This agreement carries more weight in terms of long-term roadmap advancements. After all, for Meta this is a bet involving not just selling chips but multi-generation commitments to the AMD roadmap. AMD benefits from large-scale deployment, which brings not just revenue scale but also ecosystem and software maturity.

When a hyperscaler like Meta commits billions to AMD, the implications are much further reaching than an endorsement by OpenAI and the message to the market is clear: the AI accelerator landscape is becoming multi-vendor and software ecosystems could one day evolve beyond the current CUDA dominance. For Meta, it is diversifying its compute supply and avoiding overreliance on NVIDIA. Meta needs to meet massive demand for inference workloads powering services in the next few years, as its acquisition of Manus and diversification into proprietary LLMs indicate a shift in GenAI strategy towards accelerated monetization.

This partnership is not so much about hardware but software: NVIDIA’s success is based on CUDA and its associated software ecosystem. Chip designers and makers need not just develop and manufacture the silicon, they also need to develop, test, support, and update a software stack that will require massive R&D investment and effort. For the industry, anything eroding NVIDIA’s dominance is good as it drives innovation. One of the reasons behind NVIDIA’s dominance in AI/ML was that the software stack is optimized for the workloads. But competing on the software stack represents a mammoth task ahead for AMD and is going to involve a sustained effort over a long period of time. NVIDIA chips drove the first age of GenAI compute: training LLMs. Every company except for Google, which has its own TPUs, uses NVIDIA chips for training. However, we are now in the age of inference. Although NVIDIA’s Grace Blackwell and upcoming Rubin architectures do retain an edge against AMD’s Instinct MI400 chips in training, Meta’s AI labs are now focusing on developing other areas of the business instead of frontier model development, and this is where AMD steps in.

MWC26: Telcos Vie for Limelight in Digitization Space

by Charlotte Dunlap, posted in Big Data
Close-up portrait of a smiling woman with blonde hair and earrings, set against a blurred background.
C. Dunlap
Research Director

Summary Bullets:

• Telcos are increasingly repositioning themselves as cloud and AI providers, targeting the digital transformation opportunity from an infrastructure-led perspective.

• Strategic partnerships between telcos and hyperscalers are likely to drive the most successful business transformations.

Telcos have an important agenda during this week’s high-profile technology conference, Mobile World Congress in Barcelona. They are going after the digital transformation space from an infrastructure perspective, addressing enterprises’ desperate need for improved operational provisioning, an issue largely ignored by cloud platform providers.

Telcos are working to rebrand themselves as AI, cloud, and edge providers, alongside long-established hyperscaler and platform services rivals. Operators are accomplishing this via connectivity with AI tooling (and necessary compute), leveraged through partnerships, and released as part of their evolving portfolios including edge cloud, sovereign cloud, and AI-injected connectivity. The infrastructure leaders are going after cloud giants’ customer base, including enterprises and the public sector.

The cloud ecosystem is evolving quickly and is more confusing than ever, because there’s so much competition and cooperation among players spanning the entire cloud stack. The early years of digital transformation focused on the app modernization part of digitization, now the industry is focused on infrastructure modernization. That means networking and telecommunications players are stepping up to fill in the gaps left by app platform services providers (or PaaS) such as IBM, Microsoft, AWS, Google, Oracle, SAP, and others.

What this means for enterprises is an increase in hyperscaler partnerships to enable telco workloads in the cloud including edge compute, and a trend towards telcos being able to co-sell cloud services (think Red Hat OpenShift platform services). Those platform service providers opening up their Kubernetes container and GenAI service opportunities to new infrastructure partnerships will open the floodgates of opportunity for themselves and partners. Platform providers have been faltering in their ability to keep up with integration and consulting demands from enterprise customers. Telco/platform partnerships represent a win-win for the industry.

Telcos and network providers are leveraging GenAI and even AIOps to improve their network automation capabilities and to ensure secure connectivity. In this way, managed cybersecurity represents a top-shelf digital transformation offering that infrastructure players can lead with as they go after the cloud space. Enterprises know that telcos will handle the heavy lifting of infrastructure modernization requirements (e.g., redesigning network architecture) to support these technologies.

Finally, operators are exposing network functions through standardized APIs in order to build out enterprise developer followings and play a role in the application layer of the cloud stack. Popular network services include identity, fraud detection, location, and QoS. These activities are often associated with GSMA/Open Gateway.

In many ways, MWC is more important than ever to global enterprises and the evolving role of telcos in supporting their app and infra modernization efforts. Businesses are counting on digitization to provide new business opportunities to ensure their future success.

Cloud Wars Shift to AI Wars: Archaic Code is a New Target of Opportunity

by Charlotte Dunlap, posted in Big Data
Close-up portrait of a smiling woman with blonde hair and earrings, set against a blurred background.
C. Dunlap
Research Director

Summary Bullets:

• Google will release a broadly backed commerce protocol this quarter that threatens Amazon’s behemoth e-commerce marketplace

• Anthropic Claude Code includes new features that help developers bypass the complexities of COBOL, threatening IBM’s massive mainframe business.

The cloud wars have quickly morphed into the AI wars among leading platform services rivals, which have invested heavily in LLMs in recent years. By leveraging agentic AI, platform services providers are elevating their competitive threat in their quest for dominance in both e-commerce markets and among enterprise developer communities.

GlobalData’s GenAI Quarterly report released this week (Please see Generative AI Quarterly Q1 2026: Google Goes After Amazon’s Commerce Dominance), highlights aggressive moves among cloud competitors which seriously target arch-rivals’ deeply entrenched revenue sources.

Earlier this quarter Google announced an agentic Universal Commerce Protocol (UCP), to eliminate the need for retailers to maintain complex infrastructures necessary to shore up their online purchasing lifecycles. This common language works across agents and systems to standardize most of the purchase journey including authentication, payment, and fulfillment. The alternative has been for DevOps team members to painstakingly integrate connections between various systems and partners, a major roadblock in efforts to digitize companies. Professional services required to connect infrastructure and modern applications are expensive and not always readily available. Google’s quest for a de facto agentic commerce standard is a major blow to its hyperscaler rivals, as UCP reduces the need for bespoke integration work, ultimately shifting leverage from competing platform services. It also addresses commerce partners’ strong desire to optimize AI to help drive new revenues.

In the same quarter, Anthropic revealed to its broad developer base new features within Claude Code that can modernize the legacy programming language COBOL. The language was created in 1959 but remains in use, shoring up multiple government and banking transactional processing systems, including IBM mainframes. Refactoring legacy code like COBOL represents a daunting task even among the industry’s most skilled coders. If Claude Code can significantly cut that learning curve via GenAI automation, the features represent a major new app modernization opportunity for enterprise developers to chase.

IBM X-Force Threat Index 2026: Adversaries Use AI as a Weapon in Scaling Attacks

by Amy Larsen DeCarlo, posted in Secure
A professional headshot of a woman with long blonde hair, smiling softly while wearing a black blazer and a light-colored turtleneck.
Amy Larsen DeCarlo – Principal Analyst, Security and Data Center Services

Summary Bullets:

• New research from the IBM X-Force threat intelligence team said the most sweeping developments in cybersecurity are threat actor exploiting exposed systems, gaps in supply chain defenses and fissures in interlinked application and cloud ecosystems to increase the volume and effectiveness of their attacks.

• IBM X-Force saw a dramatic rise in the number of active ransom groups, noting that cybercriminals are employing leaked tools and playbooks while using AI to automate attacks.

It is no secret that the enterprise is under threat from ambitious and aggressive cybercriminals, and that these threats have been escalating. Recently published research from IBM X-Force bears that out, highlighting the fact that adversaries are quick to exploit some major vulnerabilities to breach their targets. Compiling data from incident response, penetration tests, the dark web, and other intelligence, the newly published X-Force Threat Intelligence Index 2026 uncovered that the most common entry point for bad actors is publicly-facing applications. Citing the increasing complexity of applications and the frequency of misconfigurations, these applications are easily breached. There was a 44% increase in the number of publicly facing applications breached this year versus last.

Threat actors are polishing their techniques and applying advanced technology to infiltrate networks through vulnerabilities and other security gaps. The report suggests too many organizations are not deploying appropriate controls to deflect attacks. Fifty-six percent of the disclosed vulnerabilities did not involve authentication for access. The result is a high rate of success in stealing high-value data including credentials. IBM identified 300,000 AI chatbot credentials up for sale on the dark web.

Cybercriminals continue to actively exploit supply chain weaknesses in ecosystems, CI/CD platforms and cloud infrastructure. IBM X-Force researchers saw more attacks against developer platforms including GitHub and GitLab and breaches of cloud services infrastructure and SaaS platforms. A key takeaway is that bad actors are focusing on the platforms where applications are developed and the ecosystems that facilitate workflows.

AI is in play with adversaries employing generative AI (GenAI) to expand phishing campaigns, expedite malware development, and innovate social media campaigns with more sophisticated content creation. The immediate result is that the threat actors are using AI to increase their efficacy by reducing development time and trying out new tactics during an intrusion. This kind of agility seriously tests security practitioners who have often relied on fixed rules and signatures to combat breaches.

Conversely, enterprise security teams are also applying AI in areas like analytics to process massive volumes of network and systems data, accelerating detection and response times. Machine learning has long been a key defensive tool in discerning harmless anomalies from serious threats.

Organizations need to close security gaps, particularly at points of interconnection in their ecosystems. Fundamentals around identity and authentication and configuration issues must be addressed before they are exploited. Enterprises need to prioritize effective policy development and training, for both security practitioners and line of business employees. Security teams also need to be ready for the AI factor in attacks, which lends adversaries both speed and flexibility. At the same time, they need to tap the technology as a defensive shield in helping them recognize threats faster and deflect attacks.

The AI Boom and Unintended Consequences

by Steven J. Schuchart Jr., posted in Cloud Enable, Communicate, Secure
S. Schuchart

Summary Bullets:

• Service providers, tech companies, and enterprises are under extraordinary pressure to develop and deploy AI solutions.

• Enterprises will need to re-evaluate their goals for the rest of the year, considering cost increases and possible delivery delays.

In the late 1930s, Sociologist Robert K. Merton wrote about unintended consequences, where some of the outcomes of purposeful action are not predicted or intentional. He proposed three types of unintended consequences – unexpected benefit, unexpected drawback, and perverse result. In the first, there is a positive benefit, something good happened that wasn’t foreseen. In the second, an unforeseen detriment occurs. In the third, the result makes the problem worse, a contrarian result.

Continue reading “The AI Boom and Unintended Consequences” →

Cisco Research Finds AI Is Revolutionizing the Privacy Landscape

by Amy Larsen DeCarlo, posted in Secure
Amy Larsen DeCarlo – Principal Analyst, Security and Data Center Services

Summary Bullets:

• The overwhelming majority of organizations say they are benefiting from privacy investments in a myriad of ways, including helping increase customer trust.

• But most acknowledge data localization efforts increase the cost and complexity – and risk – to cross-border data transfers.

Data privacy is top of mind for most organizations, and not just because of a stormy geopolitical climate that is putting pressure on businesses to meet stringent regulatory requirements around data residency. Security concerns about protecting both customer data and intellectual property have companies expanding privacy programs. There is also growing recognition that customers place a premium on knowing their data will be protected.

Continue reading “Cisco Research Finds AI Is Revolutionizing the Privacy Landscape” →

Ooredoo Carves Out Fiber and Subsea Unit to Accelerate Regional Connectivity Ambitions

by Brendan Swan, posted in Cloud Enable, Communicate
B. Swan

Summary Bullets:

• Ooredoo Group launches Ooredoo Fibre Networks, an independent entity dedicated to managing and scaling Ooredoo’s international connectivity and submarine cable infrastructure ambitions.

• By establishing a new entity, Ooredoo can be financially independent from its domestic operations, enabling it to be more transparent to present and future investors.

As global bandwidth demand surges and AI-driven cloud expansion continues to reshape traditional global traffic corridors, Ooredoo’s decision to establish a standalone business dedicated to international subsea cable and terrestrial fiber connectivity signals a shift in strategic intent. Rather than simply expanding capacity, the carrier appears to be reorganizing its infrastructure to compete more assertively in the wholesale and transit arena, where scale, route diversity, and commercial agility increasingly define market leadership. In a region that is fast-emerging as a key route, is Ooredoo positioning itself to evolve from a multi-national connectivity provider into the Middle East’s regional connectivity powerhouse?

Continue reading “Ooredoo Carves Out Fiber and Subsea Unit to Accelerate Regional Connectivity Ambitions” →

Sparkle and VDPC Redefine Infrastructure Partnership; Barracuda Swims Between Italy and Spain

by Brendan Swan, posted in Communicate, IT Leader
B. Swan

Summary Bullets:

• Sparkle has teamed up with VDPC and private equity firm Teset Capital will build the Barracuda Subsea Cable, a high-capacity subsea cable connecting Spain to Italy.

• Sparkle will acquire the Valencia-Genoa infrastructure and Valencia co-location, boosting its connectivity across the Iberian Peninsula.

In today’s interconnected world, subsea cables are the unseen arteries that power global communications with underwater cables carrying approximately 99% of the world’s internet traffic across continents. As operators collaborate to expand their network infrastructure, the recent announcement between wholesale network provider, Sparkle and Spanish telecommunications provider, Valencia Digital Port Connect (VDPC) highlights how strategic collaborations between infrastructure players continue to shape the future of digital connectivity. But as ownership of Sparkle is about to change, could this influence its long-term role in global infrastructure?

Continue reading “Sparkle and VDPC Redefine Infrastructure Partnership; Barracuda Swims Between Italy and Spain” →