Also Read: Cloudflare Expands Support for AI Coding Agents via Cursor Integration

“AI can generate code fast. The harder part is getting that code safely into production. Developers should not have to leave the tools where they build software or become AI experts to put agents to work. Agentia Headless brings governed AI agents directly into the IDE and terminal. The agent does the work. The developer stays in control. That is how we turn AI-generated code into software the business can actually trust and deploy.” Highlighting early enterprise validation, Sachin Sonkar, Senior Strategy Consultant at Cognizant, added: “Early access to Agentia Headless gave us a clear picture of where enterprise Salesforce development is heading. The governance stays intact for the release manager, but the developer experience is finally catching up to how modern engineering teams actually work.” Ultimately, set for general availability in November 2026, Agentia Headless shifts AI agents from basic chat assistants into active pipeline operators, providing enterprise software teams with a seamless bridge between AI productivity and strict application lifecycle governance.

Read More: Copado Brings Governed AgentOps Directly to Salesforce Developers with Agentia Headless

The post Copado Launches Agentia Headless to Bring Governed AgentOps to Salesforce Developers appeared first on ITDigest.

In Hardware and Network news this week…

Zayo Launches Agentic Networking to Power AI Operations

Zayo is advancing network infrastructure with agentic networking capabilities designed to support increasingly complex AI operations. The approach brings AI-driven automation into network management, helping organizations respond more intelligently to changing workloads and connectivity requirements. As AI applications place greater demands on network performance, reliability, and scalability, agentic networking could help enterprises move toward more autonomous infrastructure operations while reducing manual intervention.

In Business Technology news this week…

TeamViewer Expands Tia AI Platform From IT Guidance to Governed Action

TeamViewer is expanding its Tia AI platform beyond providing IT guidance to enabling governed actions across technology environments. The development reflects a broader shift toward AI agents that can actively execute operational tasks rather than simply recommend next steps. By introducing governance into these actions, organizations can gain greater control over automation while maintaining oversight of how AI interacts with IT systems. This could help businesses improve service efficiency while addressing concerns around security, accountability, and operational control.

In Cloud Computing & Mobility news this week…

Cloudflare and OpenAI Partner to Redefine Vulnerability Management With AI-Powered Edge Defense

Cloudflare and OpenAI are collaborating to strengthen vulnerability management through AI-powered defenses at the network edge. The partnership highlights how AI can be incorporated into security infrastructure to identify and respond to emerging threats more efficiently. By bringing intelligent security capabilities closer to where traffic and applications operate, enterprises could improve their ability to detect vulnerabilities and respond to attacks while maintaining protection across increasingly distributed digital environments.

In Quantum Computing news this week…

IBM, Lockheed Martin and ETH Zurich Partner to Launch Switzerland’s First Quantum Innovation Hub

IBM, Lockheed Martin, and ETH Zurich are joining forces to establish Switzerland’s first quantum innovation hub. The initiative is expected to strengthen collaboration among industry and academia while accelerating research and practical applications of quantum computing. Such partnerships are becoming increasingly important as organizations explore quantum technologies for complex computational challenges. The new hub could also contribute to Switzerland’s position within the global quantum ecosystem by connecting research capabilities with industrial use cases.

In FinTech news this week…

Ant International, Mastercard and Visa Unveil Know Your Agent Interoperability Framework to Secure Agentic Commerce

Ant International, Mastercard, and Visa are working to establish a Know Your Agent interoperability framework aimed at making agentic commerce more secure. As AI agents increasingly participate in purchasing and financial transactions on behalf of users, financial institutions need reliable mechanisms for identifying and validating those agents. The framework addresses an emerging challenge in digital commerce by supporting greater trust and interoperability. For banks, payment providers, and merchants, such standards could become increasingly important as autonomous transactions move closer to mainstream adoption.

In HealthTech news this week…

Unchained Labs Launches AI Developability Workcell for Biologics R&D

Unchained Labs is introducing an AI-powered developability workcell designed to support biologics research and development. The solution brings automation and artificial intelligence into laboratory workflows, potentially helping researchers evaluate and optimize biologic candidates more efficiently. As pharmaceutical and biotechnology companies face pressure to accelerate discovery while managing increasingly complex research processes, AI-enabled laboratory automation could help reduce repetitive work, improve consistency, and shorten development timelines.

In Information Security news this week…

Hexnode Unveils Hexnode Synapse, Bringing Agentic Orchestration to IT and Security Operations

Hexnode is introducing Hexnode Synapse, an agentic orchestration capability aimed at IT and security operations. The platform reflects the growing use of AI agents to coordinate workflows across enterprise technology environments. By combining orchestration with IT and security operations, organizations can potentially automate routine processes, improve response times, and reduce operational complexity. The move also demonstrates how agentic AI is becoming increasingly embedded in enterprise management and security rather than remaining limited to standalone AI applications.

Article of the Week

Data Breach Incident Response: A Step-by-Step Guide for Detecting, Containing, and Recovering From Cyberattacks

Data breaches require organizations to respond quickly and systematically, from identifying suspicious activity and containing compromised systems to recovering operations and preventing repeat incidents. This guide explores the key stages of an effective data breach incident response strategy and highlights why preparation, monitoring, communication, and recovery planning are essential. As cyberattacks become more sophisticated and enterprise environments grow more distributed, having a structured response framework can help businesses limit damage, protect critical information, and restore operations more effectively.

The post ITDigest’s Weekly News Roundup Featuring Zayo, TeamViewer, Cloudflare, OpenAI, IBM, Mastercard, Visa, Hexnode and more appeared first on ITDigest.

Also Read: UST Launches UST Codon to Bring Governed AI Development Across Enterprise Platforms 

“As AI accelerates software development and vulnerability discovery, enterprises need a faster and more scalable approach to remediation. Lightwell represents a significant advancement in securing the open-source software supply chain by bringing AI-driven remediation and trusted software maintenance into the enterprise. Through our collaboration with IBM, LTM will help organisations strengthen their cyber resilience at scale.” Reaffirming the necessity of ecosystem-wide security standards, Sandip Patel, Managing Director at IBM India and South Asia, added: “Addressing today’s software supply chain challenges requires a collective defense. As AI accelerates vulnerability discovery, collaboration across the ecosystem becomes increasingly important. Bringing LTM’s engineering and transformation expertise to Lightwell can help enterprises mitigate risk and build more resilient software supply chains.” Ultimately, this collaboration combines LTM’s DevSecOps expertise with IBM and Red Hat’s AI infrastructure to accelerate secure, resilient digital innovation across complex global software ecosystems.

Read More: LTM Collaborates with IBM and Red Hat on Lightwell to Advance AI-Driven Open-Source Software Remediation

The post LTM Collaborates with IBM and Red Hat on Lightwell to Advance AI Software Remediation appeared first on ITDigest.

Security intelligence frequently sits isolated across disparate data feeds, internal files, point solutions, and legacy management platforms. In isolation, a single threat indicator offers little actionable value without understanding how it connects to a protected executive, a prior security incident, an active investigation, or a high-risk facility. Ontic AI addresses this fragmentation by unifying operational data, ensuring that generated insights carry the complete contextual backing required for rapid threat assessment and incident response.

“AI initiatives often start with the model. We believe they should start with the context of the security operation,” said Nitin Navare, Chief Product and Technology Officer at Ontic. “Ontic already connects the people, signals, incidents, investigations, and history that shape security decisions. Ontic AI builds on that foundation to make routine work easier and surface connections teams may not otherwise see.”

Also Read: Sublime Security Integrates with CrowdStrike Falcon Next-Gen SIEM

Purpose-Built Capabilities for Modern Security Workflows

Ontic AI is engineered around the core functional tasks that security analysts perform daily:

Summarize Complex Information: Transforms raw intelligence feeds, field research, entity records, incidents, and active investigations into concise, shareable intelligence overviews.

Identify Developing Patterns: Groups related signal inputs to help analysts spot emerging threat narratives and trends as they evolve.

Plain-Language Data Queries (Chat): Allows security teams to explore data repositories and extract operational insights using conversational natural language.

Reduce Repetitive Review (Deduplication): Groups redundant threat signals automatically, freeing analysts to focus on net-new security developments.

Bring Context into Triage (Copilot): Surfaces historical precedents, critical changes, and suggested remediation steps while keeping human analysts in total operational control.

“Ontic AI helps our team make sense of far more information than we could reasonably review on our own,” said the Director of Investigative Services at a Fortune 250 Business Services Company. “By bringing relevant context and patterns forward in the platform we already use, it lets our analysts focus where professional judgment matters most.”

Human-Centric Governance and Data Privacy Standards

Ontic AI is designed to augment, rather than replace, human expertise. Security analysts evaluate all AI-generated outputs and retain final decision-making authority over all security responses.

In order to stay compliant with high enterprise security requirements, we do not use client data to train or validate AI models that are not part of Ontic. Also, all AI workflows will follow the role-based permissions, data governance, and privacy settings that have been set up in Ontic already.

The post Ontic Unveils Ontic AI to Streamline Security Operations Workflows appeared first on ITDigest.

Tool sprawl has become one of the defining frictions of IT and security operations, with industry research showing that the average enterprise uses approximately 75 security tools.

“If you look at the real work of IT, nothing ever happens inside just one system. Onboarding a person touches identity, devices, and your HR system. Responding to a threat touches the endpoint, the user’s access, and the ticket. And today, a person is the one connecting all of that, every single time,” said Apu Pavithran, CEO and Founder of Hexnode, at HexCon26. 

Each handoff adds manual work, slows response and makes it harder to understand how the issue was resolved.

Hexnode Synapse connects those steps. It extends workflows across the enterprise systems teams already use, and runs deepest on Hexnode UEM, Hexnode XDR and Hexnode IdP.

Also Read: Seclore and Glean Partner to Deliver Context-Aware, Persistent Data Security for Enterprise AI

What is Hexnode Synapse?

Hexnode Synapse is a governed, agentic orchestration layer built to coordinate requests, alerts and events end to end, across every enterprise system involved in resolving them.

Synapse starts with native context from Hexnode UEM, Hexnode XDR and Hexnode IdP and extends workflows into surrounding IT and security tools. Rather than requiring administrators to move manually between consoles, it coordinates agents, approvals and system actions around a single request, alert or event.

How does Hexnode Synapse work?

“The shift is agentic: where the platform is not just telling you what to do, it’s actually doing the work,” Pavithran told the HexCon26 audience. 

Hexnode Synapse works through specialist AI agents that reason about incoming requests and alerts, determine the appropriate next step and act within defined guardrails. It initially includes:

• Device Agent: Supports root-cause diagnostics, compliance-drift remediation, patch and vulnerability remediation, and device provisioning through Hexnode UEM.
• Identity Agent: Supports access requests, identity lifecycle operations, role changes and account-related tasks through Hexnode IdP.
• Threat Agent: Supports alert triage, threat investigation and containment through Hexnode XDR.

Beyond the built-in agents, organizations can configure custom agents for scenarios around their own tools and knowledge base sources. 

Hexnode Synapse can also coordinate cross-product sequences when a workflow spans more than one system.

For example, a single prompt like “Onboard John” reads the new hire’s HR profile and kicks off the onboarding workflow: Hexnode Synapse creates the identity and assigns app licenses based on the employee’s department, pushes zero-touch device configuration and endpoint protection policies in parallel, and records the corresponding IT ticket, all as one governed workflow. If a step stalls, it recovers on its own; when app licenses run out, Synapse raises a ticket and resumes provisioning the moment a seat frees up.

How is Hexnode Synapse governed by design?

Hexnode Synapse uses guardrails, the admin-defined limits on what an agent can ever do on its own, together with action-level approval gates and a traceable execution history to keep agent actions under human control. 

Work within those guardrails proceeds autonomously. Sensitive actions, such as a device wipe, endpoint isolation or privileged access change, are routed to a named approver before execution.

“Routine work can happen on its own. Anything consequential comes back to a person, and if the system is not sure, that comes back to a person too. Autonomy without that control is not something an IT team will ever deploy,” Pavithran said.

Every autonomous or approved action is recorded in the Execution Centre: what triggered the request, what the agent decided, who approved it and what ran. IT and security teams can browse and export this history, turning an audit into a download rather than a multi-week evidence hunt.

Initial integrations and availability

At launch, Hexnode Synapse integrates with ServiceNow, Zendesk, Microsoft Entra ID and BambooHR, extending orchestration into surrounding IT and security workflows alongside Hexnode UEM, Hexnode XDR and Hexnode IdP. Hexnode plans to steadily expand this integration catalog along with Synapse’s core capabilities in the phases ahead.

The post Hexnode unveils Hexnode Synapse, bringing agentic orchestration to IT and security operations appeared first on ITDigest.

That is where data breach incident response earns its place. This isn’t merely about restoring systems. It is about learning from what happened, containing loss, preserving evidence, handling liabilities, and resuming business- without digging back into the same pit. This guide provides a six-step blueprint that outlines what companies should be doing pre, intra- and post-breach.

What Is Data Breach Incident Response and Why Is It Critical?

An ordinary IT outage can be disruptive without involving a security compromise. A server might fail; an application might crash or a network connection might go down. The immediate concern is availability. A data breach is different because the question is no longer just whether a system works. It is whether someone who should not have access has reached the information inside it.

That information could include customer records, employee details, financial information, health data or intellectual property. Once unauthorized access is suspected, the response has to answer a much harder set of questions. What was accessed? How did the attacker get there? How long were they inside? Was information copied or changed? Are other systems exposed?

NIST’s data confidentiality guidance puts the consequences plainly. Data breaches can create monetary, reputational and legal impacts, which is why its guidance focuses specifically on detecting, responding to and recovering from attacks against data.

The important point is that the breach itself is only part of the problem. The way an organization responds can determine whether a difficult incident remains contained or turns into a wider business crisis.

The 6 Essential Phases of a Data Breach Incident Response Plan

There is a reason incident response needs a plan before anything goes wrong. Once an attacker is inside, people are working with incomplete information, systems may be unstable and senior executives want answers immediately. That is a poor environment for inventing responsibilities and deciding who should make the next call.

The existing SP 800-61 Rev. 3 of NIST defines incidents response in larger cybersecurity risk management context as part of CSF 2.0. The purpose of the guidance is assist organizations in improving preparedness for, and preventing occurrence or effects of, the incidents in detection, response and recovery process.

The following six phases turn that broader approach into a practical workflow.

Also Read: Eco-Friendly Enterprise Software Initiatives: How Businesses Can Build More Sustainable Digital Operations

1. Preparation and Building Your Defense Before the Attack

The best breach response begins long before the breach.

An enterprise needs a defined incident response team with clear ownership across security, IT, legal, communications and, where needed, HR. That sounds obvious until an incident actually happens. Then small gaps become expensive. Who can isolate a server? Who speaks to customers? Who contacts external investigators? Who can approve a system being taken offline?

The same thinking applies to the technology environment. According to CISA organizations can have internet-facing assets that they aren’t even aware of which have the risk exposure. CISA’s guidance is about identifying which of those internet facing devices we do want exposed (and which are actually safe) and then minimizing those we don’t, apply patches, migrate away from unsupported systems, log traffic, use multi-factor authentication, carry out assessments and finally repeat!

Preparation then is not about creating a fancy document, it is about identifying and then minimizing things that could go wrong and somebody knows how to fix them when they do

2. Identification and Detecting the Breach Accurately

The first alert is rarely the whole story.

A security team might see an unusual login, an endpoint alert or unexpected network activity. The mistake is to jump straight from ‘something looks strange’ to ‘we have a confirmed breach.’ Investigation has to establish what actually happened.

That means checking authentication records, system and application logs, endpoint activity and network traffic. The team needs to build a timeline rather than collect isolated alerts. A suspicious login by itself may mean little. The same login followed by privilege changes, unusual file access and traffic to an unfamiliar destination tells a very different story.

Scope matters just as much. Which systems were touched? Which accounts were involved? What data could those accounts reach? Was sensitive information actually accessed?

Good data breach incident response is therefore evidence-led. Teams need enough information to make a containment decision without destroying the evidence they may need to understand the attack later.

3. Containment and Stopping the Bleeding

Once the breach is confirmed, the temptation is to shut everything down.

Sometimes that is necessary. Often it is not.

Containment should reduce the attacker’s room to operate while keeping the investigation intact. That can mean isolating an affected server, disabling a compromised account, restricting network access or separating a group of systems from the wider environment. The right move depends on what the investigation shows.

The longer-term question is what keeps the attacker from coming back while the organization works on a permanent fix. That may involve temporary access restrictions, additional monitoring or emergency patching.

NIST’s Respond guidance, updated on May 27, 2026, stresses coordination with internal and external stakeholders, analysis that supports response and recovery, and mitigation intended to prevent an incident from expanding and help resolve it.

That is an important distinction. Containment is not simply pulling a cable. It is controlled damage limitation.

4. Eradication and Eliminating the Root Cause

Stopping an attacker does not mean the attacker is gone.

Eradication is where the organization goes after the mechanism that allowed the compromise to happen and persist. Malware needs to be removed. Compromised accounts need to be secured or disabled. Vulnerabilities need to be fixed. Unwanted access paths need to disappear.

More importantly, the team has to understand why those weaknesses existed in the first place.

Suppose an attacker entered through a compromised account. Resetting that password may solve one immediate problem. But what if the account had excessive privileges? What if MFA was missing? What if nobody was monitoring the account’s unusual activity?

Those questions matter because a quick technical fix can create a false sense of closure. A system may look clean while the underlying weakness remains.

This is where a mature data breach incident response process separates itself from basic incident cleanup. The objective is not just to remove the visible threat. It is to close the path that made the intrusion possible.

5. Recovery and Restoring Operations Safely

Businesses understandably want to restore normal operations as quickly as possible. Customers are waiting. Employees cannot work properly. Revenue may already be affected.

Speed still needs a boundary.

Systems should return in stages, starting with the services that matter most to the business. Before that happens, teams need confidence that the original access path has been dealt with and that restored systems are not carrying the same compromise back into production.

Clean backups become particularly important here. So does continued monitoring after restoration. The attacker may have left behind compromised credentials, persistence mechanisms or other ways to regain access.

Recovery therefore has two jobs. It gets the business running again, and it proves that the environment is safe enough to keep running.

That second part is easy to underestimate. Restoring a system is a technical action. Restoring trust in that system is a much bigger one.

6. Lessons Learned and the Post-Incident Analysis

The final phase is where organizations decide whether the breach was merely an expensive interruption or a useful warning.

A post-incident review should not become a meeting where everyone explains why their part of the response was reasonable. That produces very little value.

The better questions are harder. Where did detection slow down? Which decision took too long? Did the right people have access to the right information? Did the escalation process work? Were there controls that existed on paper but failed in practice?

The answers should lead to changes in the incident response plan, technical controls, access policies and training.

The point is not to produce a long report that disappears into a shared folder. The point is to make the next response materially better.

Navigating Compliance and Legal Obligations During a Breach

A data breach stops being an IT-only problem the moment regulated or sensitive information enters the picture.

Legal teams need to understand what happened. Communications teams may need to prepare statements. Executives may need to make disclosure decisions. Depending on the business and jurisdiction, regulators, insurers, customers, partners or law enforcement may also become part of the response.

This gets complicated for companies operating across borders. OECD’s 2026 work on cybersecurity regulation points to growing fragmentation between jurisdictions. It says this can increase compliance costs, divert resources away from core cybersecurity functions and make international cooperation harder.

Notification should thus form a component of the response plan itself, rather than become a subject of research following the compromise. Which regulations apply and which categories of data incur notification obligations are questions the organization must understand in advance-who has the authority to approve notifications is another.

GDPR and the different state privacy statutes in the United States, for instance, cannot be easily adapted to one simple data breach response checklist because different rules apply depending on locale and information type, among other things. The Legal team must review and approve any regulatory or public notification prior to announcement to the regulatory body or the public.

Documentation is equally important. Decisions, evidence, timelines and communications should be recorded as the incident develops. When questions arrive later, a clear record is far more useful than relying on people’s memory of a chaotic day.

Expert Best Practices to Improve Your Response Readiness

Tabletop exercises are one of the simplest ways to find weaknesses before attackers do. A simulated ransomware event, compromised account or supply-chain incident can reveal problems that a written plan hides. Running these exercises twice a year can provide a useful rhythm, but the real value comes from acting on what the exercise exposes.

The technology stack should support that readiness. Zero Trust can reduce unnecessary access, while Endpoint Detection and Response can improve visibility across devices. But neither is a substitute for a tested response process.

The strongest data breach incident response programs combine people, process and technology. Remove unnecessary exposure. Know who owns the decision. Preserve evidence. Test the plan. Then fix what the test exposes.

A Breach Is Inevitable. A Disorganized Response Is Not.

The weakest incident response plans are often the ones that look impressive before anything happens. They have pages of procedures, escalation charts and security controls. Then a real incident arrives and everyone discovers the same problem. Nobody is quite sure who has the authority to act.

That is the part enterprises should worry about.

A breach will never arrive with perfect information or convenient timing. The organization will have to make decisions while facts are still emerging. The real measure of data breach incident response is therefore not how comprehensive the document looks. It is how quickly the business can turn uncertainty into controlled action.

NIST provides the framework. CISA adds practical discipline around exposure reduction. OECD highlights the growing regulatory complexity around the response. The rest comes down to execution.

A resilient enterprise is not one that assumes it will never be breached. It is one that refuses to let confusion become the second breach.

The post Data Breach Incident Response: A Step-by-Step Guide for Detecting, Containing and Recovering from Cyberattacks appeared first on ITDigest.

Also Read: Palo Alto Networks Acquires Console to Advance Agentic Security 

“As attackers use AI to move at machine speed, defenders need AI for security to detect and defend just as fast,” said Mike Nichols, general manager, Security, Elastic. “By bringing OpenAI GPT cyber models directly into Elastic Security, we can give customers additional cyber reasoning right where security work happens, without separate model infrastructure or integration overhead.” The company expects the technology to enhance investigation and remediation by combining specialized cyber reasoning with Elastic’s security data and workflows. This announcement takes further the overall partnership of Elastic with OpenAI that aims at providing advanced AI technologies to the enterprises. Security operations teams, then again, with this integration, their manual investigation tasks would get reduced. They could also respond faster to incidents and have more time to think of higher-value security decisions as the use of AI for evil purposes is the main source of the increased speed and complexity of attacks.

Read More: Elastic Brings OpenAI GPT Cyber Models Into Elastic Security to Help Defenders Investigate and Remediate Threats Faster

The post Elastic Integrates OpenAI GPT-Cyber Models into Elastic Security appeared first on ITDigest.

In Hardware and Network news this week…

Equinix Launches Fabric One to Simplify AI and Multi-Cloud Interconnectivity
Equinix has launched Fabric One to simplify connectivity across AI workloads, cloud platforms, and distributed enterprise infrastructure. The solution is designed to help organizations establish more seamless interconnections across multiple environments while improving flexibility and network management. As enterprises increasingly distribute AI workloads across clouds and data centers, streamlined interconnectivity is becoming critical to maintaining performance and scalability.

In Artificial Intelligence news this week…

PwC and Palantir Expand Strategic Alliance to Accelerate Production-Grade Enterprise AI Deployment
PwC and Palantir are expanding their strategic alliance to help organizations move AI initiatives from experimentation into production. The collaboration focuses on combining Palantir’s AI and data capabilities with PwC’s industry and transformation expertise to support enterprise-scale deployments. The partnership reflects growing demand for practical AI solutions that can deliver measurable business value while addressing governance and operational requirements.

In Cloud Computing & Mobility news this week…

HUMAIN Partners with Compal to Expand Cloud-to-Edge AI Infrastructure
HUMAIN and Compal have partnered to expand AI infrastructure spanning cloud and edge environments. The collaboration aims to support the growing computational requirements of AI applications while enabling workloads to operate closer to where data is generated. Cloud-to-edge architectures can help organizations improve responsiveness, manage data efficiently, and support AI use cases that require low latency.

In Enterprise AI news this week…

Boomi Introduces Agent Control Plane to Solve the Enterprise AI Governance Crisis
Boomi has introduced an Agent Control Plane designed to provide enterprises with greater oversight of AI agents and their activities. The technology focuses on governance, visibility, and control as organizations deploy increasing numbers of autonomous AI systems across business workflows. Establishing centralized management will be increasingly important for enterprises seeking to scale agentic AI while maintaining security, compliance, and accountability.

In FinTech news this week…

Visa Enhances A2A Protect to Help Financial Institutions Combat Real-Time Fraud
Visa has enhanced A2A Protect to help financial institutions detect and combat fraud across real-time account-to-account payments. The solution is designed to strengthen fraud prevention as instant payment systems expand and transactions move faster than traditional monitoring processes. Enhanced real-time protection can help financial institutions reduce fraud exposure while maintaining faster and more convenient payment experiences.

In HealthTech news this week…

Abcam and Genedata Collaborate to Accelerate AI-Driven Antibody Development
Abcam and Genedata are collaborating to accelerate antibody development through AI and advanced scientific data capabilities. The partnership aims to improve research workflows and help scientists work more efficiently with complex biological information. AI-driven approaches could help life sciences organizations shorten research cycles and support the development of new therapeutic candidates.

In Cybersecurity news this week…

Cognizant and CrowdStrike Unify IT and OT Security to Protect Critical Operations
Cognizant and CrowdStrike are combining IT and operational technology security capabilities to help organizations protect critical operations. The collaboration addresses the growing convergence between enterprise IT systems and operational environments, where cyber incidents can have significant operational consequences. Unified visibility and threat detection can help organizations strengthen resilience across increasingly connected infrastructures.

Article of the Week

Eco-Friendly Enterprise Software Initiatives: How Businesses Can Build More Sustainable Digital Operations

Sustainability is becoming an increasingly important consideration in enterprise technology strategy as organizations seek to reduce the environmental impact of their digital operations. Businesses can improve efficiency through optimized software, cloud resource management, automation, energy-conscious infrastructure, and responsible data practices. By embedding sustainability into technology decisions, enterprises can reduce resource consumption while improving operational efficiency and supporting broader environmental objectives.

The post ITDigest’s Weekly News Roundup Featuring Equinix, PwC, HUMAIN, Boomi, Visa, Abcam, Cognizant, and more appeared first on ITDigest.

As corporate leaders transition agentic AI models from isolated trials into production environments, disconnected legacy software, fragmented data pipelines, and API bottlenecks frequently stall progress. The MuleSoft AI Pod resolves these technical barriers by automating API construction, system integration, and security governance across multi-cloud software ecosystems.

“As organizations look to scale Agentforce and other AI initiatives across the enterprise, integration and data readiness are becoming essential,” said Roland Berthelot, Global Head of Salesforce Studio at Globant. “MuleSoft helps establish the connectivity foundation that makes those initiatives possible. With MuleSoft AI Pod, Globant is helping customers scale integration delivery while maintaining the governance, security and expert oversight enterprise adoption requires.”

Also Read: Joget Launches Agent Lab to Accelerate No-Code Enterprise AI Development

Flexible Pod Configurations and Benchmark Delivery Metrics

Engineered to match varied enterprise integration roadmaps, MuleSoft AI Pods combine autonomous AI agents with Globant enterprise architects. The agents handle structured technical tasks including initial API design, flow development, automated code testing, and legacy data mapping—while senior engineers validate system performance, verify business logic, and maintain compliance standards.

Organizations can deploy the solution across three structured delivery tiers:

Lean Configuration: Engineered to support approximately six API or integration flows per month.

Scaled Configuration: Configured to manage up to 12 integration flows per month for mid-tier environments.

Advanced Configuration: Built for high-density enterprise workloads, handling up to 24 integration flows per month.

Internal benchmarks and current deployment pilots demonstrate significant operational improvements, including:

80% Automation: Elimination of manual, repetitive integration steps through agentic automation.

15–25% Faster Time-to-Value: Accelerated project delivery schedules across complex software architectures.

35% Efficiency Gain: Improved productivity across system design, flow creation, and code validation.

Consumption-Based AI Delivery via Glob.AI Platform

The MuleSoft AI Pod is deployed through Glob.AI, Globant‘s unified technology platform designed to streamline access to specialized AI capabilities. Moving away from traditional hourly billing or per-seat licensing models, Glob.AI utilizes transparent, consumption-linked or output-based pricing—aligning technology expenditure directly with completed integrations.

The initiative expands upon Globant’s 21-year partnership with Salesforce, backed by more than 1,200 completed projects, a 4.7 out of 5 customer satisfaction rating, and over 27,400 global technologists. As a founding partner of the Agentforce Partner Network, Globant maintains over 500 Agentforce Certified Experts and 200+ Salesforce Data 360 specialists across 58 international delivery centers.

The post Globant Unveils MuleSoft AI Pod to Accelerate Enterprise Agentic AI Deployments appeared first on ITDigest.

The OECD makes this tension clear. Digital technologies can improve optimization, decision-making and energy efficiency, but they can also increase demand for electricity, water and other resources through data centers.

Eco-friendly software work should not stop at the CSR talk. The bigger chance is in everyday software choices. How a company builds, runs, and oversees systems matters? If teams use the cloud in a more careful way, and automate tasks with more thought, they can cut extra usage. Cleaner code and better data habits also help. This piece will cover where that change can begin. It will also show how IT leaders can treat sustainability like a normal way of working.

The Business Case for Sustainable IT Meets FinOps and GreenOps

Sustainability becomes much easier to defend inside a business when it stops sounding like an extra expense. Enterprise technology teams already spend significant time dealing with cloud bills, unused infrastructure, oversized workloads and resources that keep running even when nobody needs them. That is where FinOps and GreenOps begin to overlap.

FinOps focuses on getting better financial control over cloud usage. GreenOps applies a similar discipline to environmental impact. Both ask a surprisingly similar question. Are we using more resources than the workload actually needs?

That question exposes a common problem in enterprise environments. Teams often provision for peak demand, keep development environments running around the clock and accumulate workloads that nobody has reviewed for months. The infrastructure may be technically available, but that does not mean it is productive.

AWS states that most of its service charges correlate with hardware usage. Its guidance also argues that reducing resource consumption can create both cost savings and sustainability benefits. That connection matters because it changes the internal conversation. Eco-friendly enterprise software initiatives are not necessarily about choosing sustainability over profitability. In many cases, they are about removing operational waste that hurts both.

The regulatory side adds another layer. As sustainability reporting becomes more structured, companies will face greater pressure to understand their digital footprint rather than simply publish broad environmental commitments. Building measurement into IT operations now gives businesses a better starting point later.

The smarter approach is therefore not to launch a separate sustainability project and hope teams adopt it. Sustainability should become part of normal decisions around infrastructure, architecture, development and procurement.

Also Read: Cloud Security Best Practices: How Enterprises Can Protect Data, Applications and Cloud Infrastructure

4 Pillars of Eco-Friendly Enterprise Software Initiatives

1. Cloud Infrastructure Optimization and Green Hosting

Cloud migration alone does not make software sustainable. Moving an inefficient workload from an on-premise server to the cloud can simply move the problem somewhere else. The real gain comes when companies use the flexibility of cloud infrastructure to match resources with actual demand.

Right-sizing is a good starting point. An application that consistently uses a small part of a large compute instance does not need that capacity sitting idle. Teams can review utilization and adjust instance sizes based on real workloads. The same principle applies to storage. Old snapshots, duplicate datasets and unused volumes can quietly increase both cost and resource consumption.

Autoscaling takes this further. Google’s 2026 sustainability guidance recommends automated and dynamic scaling to reduce energy waste from idle or over-provisioned infrastructure. Instead of keeping capacity fixed because traffic might increase, organizations can allow infrastructure to expand and contract with demand.

Region selection also deserves more attention. Electricity does not have the same carbon intensity everywhere. When workloads can move between suitable regions, teams can consider the carbon characteristics of the electricity supplying those locations alongside latency, compliance and cost.

These practices make eco-friendly enterprise software initiatives less about buying new infrastructure and more about using existing infrastructure intelligently.

2. Intelligent Automation and AI for Resource Management

Automation can act like a control system for digital resource consumption. It watches demand, identifies patterns and adjusts infrastructure before waste becomes normal.

A simple example is the development environment. Testing and staging servers often remain active after teams have gone home. They may do nothing for hours, yet they continue consuming resources. Automated schedules can shut them down during periods when they are not required and bring them back when work begins.

The same principle can apply to production workloads. AI and machine learning can help predict traffic patterns and adjust capacity before demand arrives. That is more useful than simply adding more servers and hoping the extra capacity gets used.

There is also a broader lesson here. Automation should not only be used to make software faster. It should be used to make infrastructure more responsive.

That distinction matters for eco-friendly enterprise software initiatives. A system that automatically responds to workload changes can avoid the waste created by fixed provisioning. It also reduces the amount of manual intervention required from IT teams.

However, AI itself needs discipline. Adding an AI system to every workflow does not automatically make the operation more sustainable. The model, workload, frequency of use and infrastructure behind it all matter. The objective should be targeted automation, not AI for its own sake.

3. Energy-Efficient Green Coding Practices

Software can consume more resources simply because it has been built inefficiently. Bloated data processing, unnecessary calculations, poorly designed queries and repeated server requests all add work that infrastructure must perform.

Green software engineering starts by treating resource efficiency as a development concern. Developers can refactor inefficient legacy code, remove unnecessary processing and reduce data that applications move or store. Efficient algorithms and better architecture can often achieve more than simply adding additional computing power.

The growth of AI makes this issue even harder to ignore. Microsoft reported in June 2026 that a typical query to some of the largest and most capable LLMs used between 0.16 and 0.60 Wh of electricity, depending on factors such as query length, model and data-center specifications.

That does not mean companies should stop using AI. It means they should stop treating compute as an invisible resource.

Model selection matters. So does prompt design, request frequency and the amount of data send into a system. Smaller models may be enough for routine tasks, while more capable models can be reserved for complex work. The same thinking applies to traditional software.

Programming language choice can influence efficiency, but it should not become a simplistic battle between languages. Rust, C++, Python or Ruby can all support different workloads effectively. The bigger question is whether the software performs the required work with unnecessary computation.

That is the real purpose of eco-friendly enterprise software initiatives at the code level. Build software that does its job without making the infrastructure work harder than necessary.

4. Sustainable UX and Data Minimization

Sustainability is also a front-end problem.

Every image loaded, video streamed, API call made and piece of data transferred creates work somewhere in the system. A poorly optimized interface can therefore create unnecessary demand far beyond the user’s screen.

Image and video optimization is one obvious area. Applications do not need to deliver the largest possible file to every user. Efficient formats, compression and adaptive loading can reduce the amount of data transferred without damaging the experience.

Caching can help too. If the same information is requested repeatedly, serving it from an appropriate cache can reduce unnecessary trips back to the server. API design deserves similar attention. Combining unnecessary calls, reducing repeated requests and returning only the data an application actually needs can make systems more efficient.

Dark mode can have benefits in some device contexts, but it should not be treated as the headline sustainability tactic. Data minimization and efficient delivery have a much clearer connection to resource use.

This is where eco-friendly enterprise software initiatives become part of product design rather than just infrastructure management. A sustainable digital product should make efficiency part of the user experience without making the user think about it.

How to Implement a Sustainable Software Strategy?

The hardest part of sustainability is rarely understanding the idea. It is creating a system that makes the idea measurable and repeatable.

The first step is to audit current usage. IT leaders need visibility into compute, storage, data movement and idle resources before deciding what to change. Open-source tools such as Cloud Carbon Footprint can help teams begin measuring cloud-related emissions instead of relying on assumptions.

Commercial cloud platforms are moving in the same direction. AWS launched its Sustainability console on March 31, 2026. The service provides estimated carbon emissions using market-based and location-based methods, with information available by AWS Region, service and emissions scopes 1, 2 and 3. It also supports customizable reporting and API or SDK access.

The second step is to track Software Carbon Intensity, or SCI. The important shift is to stop measuring only infrastructure and start considering the software workload itself. Teams can establish a baseline, identify the biggest contributors and set improvement targets.

The third step is cultural. Sustainability checks should enter the DevOps workflow rather than remain inside an annual ESG report. CI/CD pipelines can include checks for major changes in resource consumption. Architecture reviews can consider efficiency alongside security, performance and cost. Engineering teams can also review idle resources as part of routine operational hygiene.

This is where eco-friendly enterprise software initiatives become sustainable themselves. The goal is not a one-time cleanup exercise. It is a repeatable operating model where teams continuously question whether a workload needs the resources it consumes.

Future Trends Shaping Green IT

The next phase of sustainable digital operations will involve more than cloud optimization.

Digital Product Passports are emerging as a way to make information about physical products, materials and lifecycle characteristics more accessible. That could strengthen the connection between software systems and broader sustainability requirements.

Serverless and event-driven architectures will also remain important. Their appeal is straightforward. Instead of keeping compute resources running continuously, applications can use capacity when events or workloads actually require it.

For eco-friendly enterprise software initiatives, this points toward a bigger shift. Sustainability will increasingly become an architecture decision made at the start of a product’s lifecycle, rather than a reporting exercise added at the end.

Conclusion

The uncomfortable truth is that many businesses do not have a sustainability problem first. They have a resource-management problem that sustainability is forcing them to confront.

Unused compute, oversized infrastructure, unnecessary data movement and inefficient workloads are not only environmental concerns. They are signs that digital operations are not being managed tightly enough.

That makes eco-friendly enterprise software initiatives far more practical than they may sound. Companies do not need to redesign every system overnight. They can begin by auditing cloud usage, identifying idle capacity, measuring software impact and making efficiency part of engineering decisions.

The strongest sustainability strategy is therefore not the one with the biggest promise. It is the one that quietly removes waste from everyday operations while making the business faster, leaner and easier to manage. Start with the cloud bill. The environmental footprint may be hiding in the same place.

The post Eco-Friendly Enterprise Software Initiatives: How Businesses Can Build More Sustainable Digital Operations appeared first on ITDigest.