With a broad set of tools and services for software developers, startup teams, and ISVs, the new cloud infrastructure suite will allow teams to host, deploy, and scale self-optimizing autonomous AI agents with managed vector data management, persistent memory, and continuous feedback ingestion.

Overcoming Technical Bottlenecks in Autonomous AI Agent Deployment.

As enterprise AI shifts from simple chatbot interfaces to autonomous software agents that perform multi-step actions, engineering teams face infrastructure challenges. Preserving agent context over multiple sessions, supporting high-velocity vector databases, and providing continuous feedback systems needs building on disparate third-party integrations that can introduce complexity to architecture and result in unknowable cloud costs.

Also Read: Anaconda and Intel Deepen Alliance to Establish Secure, Enterprise-Grade AI Deployment Framework

DigitalOcean’s Agent Learning and Data Services solve these operational challenges by embedding managed data infrastructure directly into its cloud ecosystem. The solution provides autonomous software agents with persistent memory and structured feedback loops, allowing them to evaluate task outcomes, adapt execution strategies, and improve performance over time without needing constant manual intervention or continuous model retraining.

Key operational capabilities and technical features include:

Persistent Multi-Session Memory: Delivers fast vector storage and retrieval to preserve contextual state across complex, ongoing user and agent interaction threads.

Managed Data & Ingestion Pipelines: Automates data chunking, embedding generation, and vector indexing to keep knowledge bases updated with live application inputs.

Recursive Performance Telemetry: Establishes structured evaluation and feedback channels, allowing autonomous software agents to analyze past actions and optimize decision-making accuracy recursively.

Cost-Optimized Developer Cloud: Deploys natively on DigitalOcean’s compute and GPU infrastructure, offering predictable, simple pricing suited for scaling AI startups and developers.

“The question has changed from ‘can we launch an agent’ to ‘does our agent get better with every run,'” said Vinay Kumar, Chief Product and Technology Officer, DigitalOcean. “We built Inference Engine so agents could think, Managed Agents so they could act, and now Agent Learning Services and Agent Data Services so they can improve, on one open cloud and on the customer’s own data. That is what owning your intelligence means, and it is why DigitalOcean is the place to launch and scale self-improving agents.”

Accelerating Scalable Cloud Innovation for Modern Developers

Using persistent memory combined with managed vector data in a developer-first cloud, DigitalOcean eliminates the largest barriers to enterprise-grade AI innovation. This implementation offers software engineers a simple, straightforward path to build, iterate, and deploy self-learning autonomous applications at scale.

The post DigitalOcean Introduces Agent Learning and Data Services to Support Recursively Self-Improving AI Systems appeared first on ITDigest.

As Chief Marketing Officer, Minnick will oversee global marketing strategy, brand development, product positioning, and go-to-market execution. His leadership will focus on driving enterprise demand generation and positioning Upwind within the competitive cloud-native application protection platform (CNAPP) and runtime security markets.

Also Read: Ontinue Expands External Threat Visibility with ION for Dark Web Monitoring

Upwind provides a cloud security platform that utilizes runtime data to defend cloud architecture, containerized environments, and cloud-native infrastructure against threats. By integrating real-time operational context into threat detection and posture management, the company enables enterprise security and DevOps teams to prioritize critical risks and reduce incident response times.

The addition of Minnick to the executive team reflects Upwind’s focus on scaling customer acquisition and broadening its industry footprint. As enterprises continue to expand multi-cloud architectures and integrate AI-driven development workflows, Minnick’s go-to-market oversight will play a strategic role in advancing Upwind‘s brand messaging and commercial growth strategy.

The post Upwind Appoints Joel Minnick as Chief Marketing Officer appeared first on ITDigest.

To help overcome this bottleneck in cloud security, the worldwide leader in information security and defense Thales unveiled a strengthened strategic partnership with Google Cloud.

The partnership integrates the Thales AI Security Fabric natively into Google Cloud Gemini Enterprise, establishing integrated runtime protection, threat detection, and policy enforcement across agentic AI workflows.

The News: Real-Time Runtime Enforcement and Agent-to-Agent Governance

The primary technical advancement delivered by the Thales and Google Cloud collaboration is the transition from perimeter-based cloud security to in-line, runtime governance for autonomous agents. Rather than treating AI models as simple software endpoints, the integrated architecture continuously monitors communications between autonomous agents, foundational models, enterprise data lakes, and third-party tools.

Also Read: Arcadis Deepens Autodesk Collaboration to Accelerate AI-Led Infrastructure Delivery

Key architectural capabilities introduced in the announcement include:

Real-Time Agentic Threat Mitigation: Protects against emerging AI vulnerability vectors including direct and indirect prompt injection, sensitive data spillages, unsafe model outputs, and unauthorized system commands.

Agent-to-Agent Communication Control: Implements granular permission boundaries governing how autonomous agents interact with one another, preventing cascading unauthorized actions across multi-agent workflows.

Integrated Gemini Enterprise Governance: Connects the Thales AI Security Fabric directly into Google Cloud’s enterprise AI suite, giving security operations (SecOps) teams unified visibility over model usage and data access.

Centralized Compliance and Audit Tracing: Generates immutable, real-time audit trails detailing agent decisions, tool execution histories, and data touchpoints to satisfy regulatory governance frameworks.

Transforming the Enterprise Cybersecurity & Cloud Infrastructure Industry

The joint release by Thales and Google Cloud marks a decisive structural turning point across the Enterprise Cybersecurity, Cloud Infrastructure, and AI Governance sectors.

The Phase-Out of “Static” Cloud Security Models
For years, cloud security platforms relied on static identity and access management (IAM) roles, API gateways, and perimeter firewalls. However, autonomous agents do not follow predictable, linear code paths; they dynamically evaluate goals, call external APIs, and execute database queries on the fly.

This integration accelerates the obsolescence of static perimeter defenses. The cybersecurity industry is entering an agentic runtime security era, where governance tools are evaluated on their ability to inspect model reasoning streams, evaluate context in real time, and intercept unauthorized agent commands before execution occurs.

Establishing “Identity for AI Agents” as a Security Baseline
As enterprises deploy hundreds of specialized AI agents across internal departments, defining agent identity and operational boundaries has emerged as a major challenge.

By embedding the Thales AI Security Fabric into Google Cloud’s AI infrastructure, the partnership establishes zero-trust agent identity as an enterprise standard. Security platforms must now treat AI agents as distinct digital identities equipped with specific least-privilege access boundaries, continuous authentication, and strict action limits.

Broad Operational Impact on Enterprise Businesses Operating in this Sector

These are the direct business cases and strategic benefits that integrated agentic security tools offer to CISOs CTOs cloud architects managing AI pipelines:

Accelerated Transition from AI Pilots to Production: Strengthening security guardrails allows executive leadership to implement autonomous agent initiatives beyond R&D sandboxes and into core business processes.

Prevention of Catastrophic Operational Anomalies: Identifies unwanted agent actions before they execute rogue commands that would change production database, delete cloud infrastructure or dump proprietory intellectual property.

Simplified Multi-Jurisdictional Regulatory Compliance: centralized audit logs of agent-driven data access then also make it easier to report toward compliance with strict privacy laws and new international AI safety regulations.

Lowered Total Cost of Security Operations (TCO): Placing agent runtime security on the core cloud platform removes the requirement for developing costly, custom in-house security wrappers for every individual AI model deployment.

By replacing reactive security measures with proactive, runtime agent governance, Thales and Google Cloud provide a viable blueprint for the agentic era enabling enterprises to unleash autonomous AI productivity without compromising systemic security or regulatory compliance.

The post Safeguarding Autonomous Execution: Thales and Google Cloud Partner to Secure Agentic AI Workflows appeared first on ITDigest.

The release arrives as enterprise AI transitions rapidly from early-stage experimentation to production-grade deployments. On Fastly’s global network, machine-generated traffic surpassed 50% in July and August 2026, while AI traffic grew 6.5 times faster than human-generated requests between January and May 2026. Simultaneously, corporate cost pressure is escalating; industry studies from McKinsey indicate that 93% of organizations are currently exceeding their allocated AI budgets. Fastly’s programmable edge infrastructure directly addresses these cost and security risks by applying real-time policy enforcement inline across public, private, and self-hosted model providers.

“To effectively adopt coding agents and implement AI features without losing velocity or eroding existing business resilience, enterprises need control in production, at runtime, without delay, friction, or disruption,” said Kelly Shortridge, Chief Product Officer at Fastly. “Fastly is extending the real-time control customers already trust for content delivery and software security to the models, applications, and agents shaping this new era of distributed systems.”

Also Read: Abnormal AI Unveils AI Cloud Security Platform with OpenAI Daybreak Models 

Three Core Critical Paths for Enterprise AI Governance

Fastly’s updated portfolio establishes centralized governance across the three primary operational vectors of modern AI systems: applications making model calls, users and external systems interacting with AI-powered applications, and autonomous software agents querying corporate APIs.

The newly released solutions include:

AI Runtime Control (Governing AI Access and Usage): Consolidates model calls through a single edge endpoint across public and self-hosted environments. Utilizing virtual keys to safeguard underlying provider credentials, the system offers real-time tracking of token spend, rate limiting, budget caps, and automatic failover routing to maintain model choice while capping runaway costs.

AI Firewall (Protecting AI Applications): Delivers inline threat protection at the edge to mitigate Large Language Model (LLM) vulnerabilities, including prompt injection attacks. By evaluating prompts directly within the request path before they reach downstream models, the firewall blocks malicious payloads deterministically without requiring GPU overhead.

API Security (Controlling Agent Interactions): Enforces strict API schemas and contract rules across agentic, AI-assisted, and conventional traffic. IT security teams can observe or block non-conforming requests on a service-by-service basis, preventing autonomous agents from initiating unsupported operations or transmitting malformed data.

Unifying Security and Edge Infrastructure

Built onto Fastly’s high-capacity global network which operates at 622 Tbps of capacity and processes over five trillion requests daily the new AI tools integrate seamlessly with the company’s existing security platform, including Next-Gen WAF, Bot Management, ContentGuard, DDoS Protection, and API Discovery. By embedding AI governance directly into the edge delivery path, Fastly enables enterprises to protect their digital assets, optimize model performance, and prevent unbudgeted expenditure from a single management console.

The post Fastly Launches AI Firewall and AI Runtime Control to Secure and Scale Enterprise AI appeared first on ITDigest.

As enterprise data lakes evolve into mission-critical lakehouses powering AI models and real-time business intelligence, open table formats like Apache Iceberg have become essential infrastructure. However, a key vulnerability has persisted: native Iceberg snapshots are merely metadata pointers, not full backups. If a table is corrupted by ransomware, deleted by an errant script, or overwritten by a misfiring AI agent, those pointers vanish. Restoring underlying storage files leaves organizations with raw, disorganized Parquet files requiring days of manual metadata reconstruction before analytics engines can query the data.

Rubrik’s new capability closes this vulnerability by backing up both data files and table metadata, automatically re-registering tables in the catalog upon recovery. This allows data teams to restore complete, immediately queryable tables in engines like Amazon Athena, Apache Spark, or Trino.

Also Read: Airties Debuts “Aura” Agentic AI Engine to Transform Connected Intelligence for ISPs

Technical Capabilities: Metadata Rewiring and In-Account Sovereignty

Coverage extends across AWS Glue Data Catalog and Amazon Simple Storage Service (S3) Tables. Key technical highlights include:

Iceberg-Aware Metadata Recovery: Captures full table definitions, history, and schema state alongside raw data. Upon restore, Rubrik re-wires the catalog automatically so analytics engines can resume operations without manual file repair.

In-Account Data Sovereignty: it stands out as the first Iceberg-aware protection solution that enables enterprises to keep immutable backups locked inside their own AWS accounts, satisfying data compliance and sovereignty requirements, and complementing Rubrik’s air-gapped Cloud Vault.

Petabyte-Scale Forever-Incremental Backups: makes use of forever-incremental method of compacted snapshot backups which keeps backup window low for a very large scale deployment and lets the companies select their preferred Amazon S3 storage tier to control cost.

Risk Validation Based on branch: gives ability to run backups right before the risky activities like schema changes or large batch writes by having the ability to restore data on a separate Iceberg branch for validation before promotion production environment.

Transforming the Cloud Data Security, Lakehouse, and AI Infrastructure Industry

Rubrik’s release accelerates a fundamental evolution across the broader Cloud Data Security, Data Management, and AI Infrastructure sectors.

The Obsolescence of “Storage-Only” Cloud Backups
Historically, cloud backup providers treated object storage (like Amazon S3) as simple buckets of unstructured files. However, modern analytical workloads rely on abstraction layers catalogs, schema definitions, and metadata pointers to make raw data usable.

This announcement highlights the inadequacy of storage-only backups for lakehouses. The data security market is shifting toward metadata-aware cyber resilience. Security vendors can no longer claim cloud data protection simply by backing up raw bytes; they must protect the catalog and logical table structures that give that data operational meaning.

De-Risking Autonomous AI Agents on Data Lakes
As enterprises deploy autonomous AI agents with write access to internal data lakes for automated ETL, feature engineering, and reporting, the risk of “agentic data corruption” where an AI agent incorrectly modifies schema or overwrites table states at scale has grown significantly.

By introducing Iceberg-aware recovery and isolated branch testing, Rubrik establishes a baseline safety layer for AI-driven data pipelines. Data engineering teams gain a critical safety net, allowing them to grant AI agents broader operational permissions without risking permanent lakehouse corruption.

Operational Impact on Enterprise Businesses

For top enterprise executives responsible for data lakes, mission-critical analytics, and AI on AWS – like CIOs, CISOs, and CDOs – adopting lakehouse-aware protection has numerous operational benefits:

No more Downtime and Lost Revenue
If ransomware attacks or deletions cause company data lakehouses to stop functioning, dashboards for business intelligence, real-time supply chain simulations, and consumer apps will all become inoperable. Shortening the recovery period from many days of manual reconstruction work to just a few minutes of automated catalog re-registration ensures the smooth operation of business as well as the avoidance of possible losses in revenues.

Compliance made Easy by Securing Data Sovereignty
Organizations like financial services, healthcare providers, and public institutions are at a regulatory risk when their sensitive data is moved for cloud backup as they rely on third-party backup solutions that usually require leaving data in the enterprise clouds. Granting these institutions the ability to keep their immutable and air-gapped backups on the same AWS tenant that they use will ensure full mastery over where they keep their data and how to encrypt it, even while cyber insurance and other compliance requirements get satisfied.

The post Rubrik Launches Lakehouse Protection for Apache Iceberg Data on AWS to Close the Cloud Data Resilience Gap appeared first on ITDigest.

The platform launch addresses an emerging class of operational risk brought into focus by the recent OpenAI-Hugging Face incident. During an internal security evaluation, autonomous AI agents identified unexpected configuration paths that extended beyond their designated environments, ultimately accessing external production infrastructure. The event demonstrated how fast-moving AI agents can uncover operational gaps, chain vulnerabilities together, and execute cloud actions at speeds that surpass traditional, human-dependent security operations.

As Greg Brockman, President and Co-Founder of OpenAI, recently wrote in his blog post “The Defender’s Window”: “The OpenAI-Hugging Face incident was a watershed moment for cybersecurity because it gave a peek into how the capabilities of a typical threat actor will evolve in upcoming months. I’ve spoken with many organizations over the past few weeks, and one theme is clear: they know they need to fundamentally uplevel their cybersecurity practices with unprecedented speed.”

“The Hugging Face incident was a warning shot. It showed what capable agents can already do when they go off-script, and why security teams need to prepare for similar techniques being used intentionally by attackers,” said Evan Reiser, Founder & CEO of Abnormal AI. “Our platform already detects behavioral anomalies from human and non-human identities and I’m excited for us to extend these capabilities to protect enterprise cloud environments.”

Also Read: Harness Launches AppSec Alliance to Secure Application, API, and Agent Workflows

Machine-Speed Defense Architecture for Enterprise Cloud Infrastructure

Legacy cybersecurity tools that rely primarily on static threat signatures struggle to identify non-standard, agent-driven exploits that lack historical precedent. Abnormal AI Cloud Security counters these gaps by applying continuous behavioral analysis—establishing baseline operational profiles for every human identity, service account, API key, and autonomous AI agent across an organization’s digital footprint.

Key architectural capabilities of Abnormal AI Cloud Security include:

Real-Time AI Breach Detection: Continuously evaluates behavioral signals across the cloud estate, flagging suspicious activity the moment an identity or agent strays from established operational norms.

Autonomous Incident Containment: Delivers machine-speed response workflows when threat conditions are met. Security administrators can configure the platform to automatically isolate compromised workloads, revoke API tokens, and contain affected resources based on incident severity.

OpenAI Daybreak Investigation Engine: Utilizes OpenAI Daybreak models to assist security teams during incident triage. The system correlates raw telemetry logs with behavioral context to map out incident blast radiuses and suggest remediation paths.

Bridging the Operational Gap in Agentic Security

By extending behavioral analytics natively into cloud workloads, Abnormal AI gives enterprise security operations centers (SOCs) the toolset needed to govern both human and software-driven actors. The platform allows security teams to maintain high-velocity cloud workflows while deploying automated guardrails that prevent autonomous agents from breaching operational boundaries.

The post Abnormal AI Unveils AI Cloud Security Platform with OpenAI Daybreak Models appeared first on ITDigest.

Addressing this escalating crisis, connectivity cloud leader Cloudflare, Inc. announced a landmark integration with OpenAI to launch Vulnerability Discovery and Remediation. Available in early access through Cloudflare Managed Defense, the new service combines frontier AI models including OpenAI’s GPT-5.6 Cyber via the OpenAI Daybreak Defense Network with real-time traffic intelligence across Cloudflare’s global edge network.

The collaboration introduces a proactive defense layer that maps active production traffic directly to source code vulnerabilities, allowing enterprises to block live exploits at the edge in seconds while automated AI agents draft verified code patches.

Technical Performance: Linking Live Traffic Signals with Frontier AI Code Analysis

The core technological advance of the service is its ability to eliminate the disconnect between static source code scanning and dynamic network activity. By analyzing source code alongside real-time internet traffic patterns, the architecture enables defenders to prioritize and neutralize threats before exploit payloads reach core infrastructure.

Also Read: PwC and Palantir Expand Strategic Alliance to Accelerate Production-Grade Enterprise AI Deployment

Key capabilities of the integrated platform include:

Context-Aware Threat Triaging: Correlates live web traffic and attack patterns with source code structures, immediately highlighting vulnerabilities that are actively under fire.

Instant Edge Protection (Virtual Patching): Generates and deploys custom Web Application Firewall (WAF) mitigation rules tailored to specific attack vectors. This shields applications at the edge in seconds while developers work on permanent software fixes.

AI-Assisted Patch Generation: Leverages OpenAI’s GPT-5.6 Cyber model to investigate codebases and automatically draft context-aware code patches for developer review.

Human-in-the-Loop Governance: Enforces explicit human approval for both WAF rules and code modifications, ensuring security teams retain total control over production deployments.

“If your security team is manually fighting AI-driven attacks, you’re not just burning them out you’re losing,” stated Matthew Prince, co-founder and CEO of Cloudflare. “Pairing Cloudflare’s global network with OpenAI GPT-5.6 Cyber shifts defense from chasing patches one by one to stopping attacks before they land.”

Transforming the Enterprise Cybersecurity and Cloud Infrastructure Industry

Cloudflare and OpenAI’s partnership signals a defining structural pivot across the Cybersecurity, Application Security, and Cloud Edge Infrastructure market.

The Obsolescence of Standalone Vulnerability Scanners
For years, the application security market relied on static application security testing (SAST) and vulnerability scanners that produced overwhelming lists of theoretical flaws.

The launch of edge-based AI remediation marks the decline of passive scanning tools. The cybersecurity industry is transitioning toward unified threat intelligence platforms where network telemetry, edge computing, and AI code analysis operate in a continuous feedback loop. Security vendors will no longer be evaluated on how many bugs they surface, but on how fast their platforms apply virtual patches at the edge.

Establishing Frontier AI as a Core Defensive Engine
Historically, security teams worried that generative AI models would primarily benefit threat actors by accelerating zero-day discovery.

By deploying GPT-5.6 Cyber natively within the OpenAI Daybreak Defense Network, Cloudflare and OpenAI establish a benchmark for AI-driven cyber defense. Enterprise buyers will increasingly expect security infrastructure to feature specialized, defensive AI models capable of autonomous code reasoning and real-time mitigation drafting.

Broad Operational Impact on Enterprise Businesses

For corporate leadership, Chief Information Security Officers (CISOs), and Chief Technology Officers (CTOs) facing surging software vulnerability rates, deploying AI-powered edge defense offers direct strategic and commercial advantages:

Shielding Revenue and Preserving Developer Velocity
To fix a critical vulnerability, usually engineering teams are forced to abandon product feature development in favor of writing, testing, and deploying emergency hotfixes. Virtual patching at the edge can instantly block exploits and protects enterprise applications without having a negative impact on product development.

Reducing Breach Liabilities and Alert Fatigue
Closed vulnerable items in a company might result in significant monetary penalties, disruptions caused by ransomware, and also damage to a company’s reputation. Filtering threats using real-time network context will help SOC teams identify the real problems, cut out alert fatigue, and thereby reduce overall cyber risk exposure.

The post Cloudflare and OpenAI Partner to Redefine Vulnerability Management with AI-Powered Edge Defense appeared first on ITDigest.

That is the part of cloud security that often gets missed. The issue is not simply protecting a server or encrypting a database. It is controlling who can access what, understanding which responsibilities still sit with the enterprise, spotting weaknesses before attackers do, and keeping security controls intact as the environment changes.

This guide breaks down the practical cloud security best practices enterprises need across identity, data, threat detection, compliance, APIs, and containers.

The most effective cloud security best practices include enforcing identity management, encrypting data, continuous threat detection, and maintaining strict compliance.

The Foundation of Cloud Security Is the Shared Responsibility Model

One of the easiest mistakes to make after moving to the cloud is assuming that the provider now owns security.

It does not work that way.

AWS draws a pretty clear line between the whole security of the cloud and security in the cloud really matters. AWS is on the hook for securing the underlying infrastructure that powers their cloud services. Meanwhile the customer stays responsible for the services and all related configurations, also for the applications, the data, and the workloads that get placed into that environment.

That distinction becomes even more important when the type of cloud service changes.

With IaaS, the customer has a larger security responsibility because it controls more of the environment. Operating systems, applications, network settings, identities, and configurations can all sit within the customer’s area of control. With PaaS, the provider manages more of the underlying platform, but the customer still has to secure applications, data, identities, and configurations. SaaS moves more infrastructure responsibility to the provider, yet the customer still controls things such as user access, permissions, and data handling.

So the shared responsibility model is not just something to understand during a cloud migration. It should influence the security architecture from the start.

There is a practical reason for this. When responsibility is unclear, security gaps tend to fall between teams. The cloud provider assumes the customer is handling something. The customer assumes the provider is handling it. The control ends up belonging to nobody.

That is a bad place for a security control to live.

Identity Management and Access Controls Put Identity at the Center

The idea of a fixed security perimeter has become much harder to maintain. Employees work remotely. Applications talk to other applications. APIs connect services. Cloud workloads can move between environments. Automated processes can access sensitive resources without a person initiating every action.

Identity sits in the middle of all of this.

A useful way to think about modern cloud access is to stop asking whether someone is ‘inside’ the network and start asking whether that particular request should be trusted. Who is requesting access? What are they trying to reach? Why do they need it? How sensitive is the resource? Does the request look normal?

That is the thinking behind Zero Trust.

Multi-Factor Authentication is an important starting point because a stolen password should not automatically become a valid entry ticket. Microsoft’s current Azure identity guidance states that MFA can block more than 99.2% of account compromise attacks.

That is a strong reason to make MFA mandatory, particularly for privileged and sensitive accounts. But authentication is only one part of IAM.

Enterprises also need Role-Based Access Control and Conditional Access so permissions can reflect actual responsibilities and circumstances. A finance employee should not automatically have access to production databases simply because the account exists. A developer may need access to production for a specific task without needing unrestricted control over the entire environment.

Least privilege is where this becomes practical. Give an identity the access it needs. Nothing more.

That rule also applies to service accounts, API keys, workloads, and other non-human identities. These identities are easy to overlook because there is no employee sitting behind them. Yet they can still hold powerful permissions and connect directly to critical systems.

A mature IAM strategy therefore looks beyond employees. It keeps track of every identity that can reach the cloud environment and regularly asks whether that access still makes sense.

Data Protection Requires More Than Encryption

Encrypting cloud data is important. Treating encryption as the entire data protection strategy is not.

Before an enterprise decides how to protect data, it needs to understand what it actually has. Sensitive information can sit in databases, object storage, backups, application logs, development environments, and temporary workloads. It can also move between services without anyone thinking of that movement as a separate security event.

That is why data protection needs several layers.

Data at rest needs appropriate encryption. Data moving between systems needs protection in transit. Access to both needs to be controlled. Encryption keys need to be managed separately from the data they protect. Secrets such as passwords, tokens, and API keys also need secure storage rather than being left inside application code or ordinary configuration files.

AWS’s 2026 VPC Encryption Controls capability is a useful example of how this is becoming more operational. It can monitor encryption status across traffic flows, identify resources that unintentionally allow plaintext traffic, enforce encryption across supported network paths, and generate audit logs for compliance and reporting. AWS says supported traffic paths use hardware-based AES-256 encryption.

The important takeaway is not simply ‘use encryption.’ It is ‘keep checking whether the intended protection is actually in place.’

Cloud environments change too quickly for a security setting to be treated as permanent. A network rule changes. A new service is connected. A workload gets moved. A configuration is copied into another environment. The original security decision may no longer hold.

Backups need the same level of attention, honestly. The automated backups help a lot, but enterprises should also think about who gets access to them, how they are safeguarded, how frequently a restore is actually tested, and if they still be restorable when the primary environment is, say, completely out of action.

Data protection is therefore a process, not a switch.

Also Read: Challenges of Managing Hybrid Cloud Environments: How Enterprises Can Improve Security, Visibility and Performance 

Proactive Threat Detection and Response Changes the Security Game

Cloud infrastructure can change faster than a security team can manually inspect it.

A new workload can appear. A permission can be changed. An application can expose a service. A developer can deploy a new container. None of those actions necessarily wait for a security review.

That creates a problem for a purely reactive approach. If the team only starts looking after an alert arrives, the weakness may already have been sitting in the environment for some time.

This is where proactive cloud security becomes important.

Cloud Security Posture Management helps teams, basically spot misconfigurations and those security weak spots across different cloud environments, yeah. Cloud Workload Protection Platforms add a more grounded view into workloads plus runtime activity. Then there is agentless vulnerability scanning, which can also help uncover weaknesses without forcing an agent installed everywhere, so it’s kind of flexible.

The technology matters, but the bigger change is how security teams deal with the information they find.

AWS Continuum, launched in June 2026, is designed to discover, prioritize, validate, and remediate security risks at machine speed. It can take findings from existing tools, use environmental and business context to determine priority, validate exploitability, and apply mitigations within defined guardrails.

That matters because a long list of security findings does not automatically mean an organization understands its risk.

A critical vulnerability on an exposed production workload deserves a different response from a low-risk issue buried inside an isolated development environment. Without context, security teams can spend their time clearing alerts instead of reducing meaningful exposure.

Good cloud security best practices therefore need a clear path from detection to action. Find the weakness. Understand its context. Work out whether it can actually be exploited. Then decide what needs to happen.

Security teams do not need more noise.

They need better decisions.

Cloud Compliance Needs Continuous Governance

Compliance is often treated as an event. The audit happens, evidence gets collected, gaps are documented, and everyone moves on.

Cloud environments do not behave like that.

Configurations change every day. New resources are created. Permissions are updated. Applications are deployed. Infrastructure is removed and rebuilt. A control that was working when an audit took place may not be working months later.

That is where configuration drift becomes a real problem.

Enterprises need automated checks that continuously compare the environment against defined security and compliance requirements. The exact requirements will vary. A healthcare organization may have to address HIPAA. An organization handling personal data may have GDPR obligations. Others may work against SOC 2 requirements or additional industry controls.

Microsoft’s Cloud Security Benchmark v2 shows how quickly the control landscape itself is expanding. Microsoft says its policy-based control measurements have grown from more than 220 to more than 420. The preview also adds a dedicated AI Security control domain with seven recommendations.

The number is useful, but the bigger message is what sits behind it. Cloud environments are becoming more complicated, and governance has to keep up.

That means compliance should connect directly to technical controls. Policies should be measurable. Changes should be visible. Exceptions should have owners. And when a configuration moves away from the expected state, the organization should know about it without waiting for the next audit.

The goal is not to make the cloud look compliant on audit day.

The goal is to keep it controlled on every other day too.

Modern Cloud Security Must Reach APIs and Containers

Cloud-native development has made applications more flexible, but it has also multiplied the places where security problems can appear.

Microservices communicate through APIs. Containers can be created and replaced rapidly. Kubernetes can manage large numbers of workloads across constantly changing environments. A vulnerable dependency, exposed API, leaked secret, or poorly configured container can therefore become a problem before the application reaches production.

Security cannot be bolted onto that process at the end.

This is where DevSecOps becomes useful. Security checks should become part of the CI/CD pipeline. Code can be checked for weaknesses. Dependencies can be reviewed. Container images can be scanned. Infrastructure configurations can be tested. Secrets can be identified before deployment.

The advantage is simple. Developers can fix problems while they are still close to the code and configuration that created them.

That does not mean runtime protection becomes unnecessary. It means runtime security no longer has to carry the entire burden.

The strongest approach combines both. Prevent what can be prevented during development, then monitor what reaches the cloud and respond when something changes.

Conclusion and Next Steps

The hardest part of cloud security is not buying another security tool. It is keeping the basics under control when the environment refuses to stay still.

Identities change. Workloads move. Permissions grow. APIs multiply. Configurations drift. New services appear. That makes cloud security an ongoing management problem rather than a one-time implementation project.

Enterprises should therefore assess security as one connected system. Review identities and permissions. Check encryption and secrets. Test backups. Monitor workloads. Look for configuration drift. Secure APIs and containers. Keep compliance controls tied to the actual environment.

A comprehensive cloud security posture assessment is a sensible place to start. A practical checklist or assessment framework can expose the gaps that individual tools often miss.

The real test is not whether the cloud was secure when it was first deployed.

It is whether the security still holds six months later.

The post Cloud Security Best Practices: How Enterprises Can Protect Data, Applications and Cloud Infrastructure appeared first on ITDigest.

In order to address these visibility issues and help companies secure their cloud transformations, digital engineering giant EPAM Systems entered into a strategic partnership with cloud security specialist Wiz.

The combination of EPAM’s global capabilities in digital engineering, cloud migrations, and managed security, with Wiz’s Cloud-Native Application Protection Platform (CNAPP) solution allows for achieving full-stack visibility, reducing cloud risk, and enhancing cyber resilience in complex cloud environments.

The News: Unifying Cloud Security Posture with Scalable Engineering

The core value proposition behind the EPAM and Wiz alliance is bridging the gap between security visibility and engineering execution. Identifying a security flaw inside a multi-cloud network is only half the battle; enterprise security teams frequently struggle to remediate risks without interrupting live production software or developer workflows.

Also Read: FS Launches 800G Muxponder to Power Next-Gen AI and Cloud Networks

The joint solution addresses these operational challenges through several key capabilities:

Full-Stack Contextual Visibility: it makes use of Wiz’s agentless CNAPP platform to construct a single, queryable risk graph that spans an organization’s complete cloud presence including virtual machines, Kubernetes clusters, serverless functions, identity permissions, and sensitive data stores.

Prioritized Risk Mitigation: it assesses potential attack vectors based on how a mix of misconfigurations, exposed secrets and unpatched vulnerabilities come together, leading to prioritization of truly dangerous threats instead of background noise.

Automatic Remediation Pipelines: With Wiz’s real-time threat detection coupled with the custom IaC coding work of EPAM, one can automate policy enforcement as well as vulnerability patching quickly and safely within CI/CD pipelines of developers.

Continuous Compliance & Governance: by setting up automated compliance monitoring through the major regulatory standards like SOC 2, ISO 27001, HIPAA and GDPR public and hybrid cloud environments will always remain audit-ready.

Transforming the IT Consulting, Cloud Engineering, and Cybersecurity Industry

The collaboration between EPAM and Wiz reflects fundamental operational shifts across the IT Services, Cloud Engineering, and Managed Security ecosystem.

The Shift from “Point-Solution Scanning” to End-to-End CNAPP Platforms
For years, IT service providers and enterprises relied on fragmented point tools—using separate products for container security, cloud security posture management (CSPM), identity analysis, and vulnerability scanning. Managing dozens of isolated scanners generated overwhelming alert fatigue while failing to show how different risks combined to create exploitable attack paths.

EPAM’s integration of Wiz highlights the obsolescence of isolated point tools. The cybersecurity and cloud consulting industry is pivoting rapidly toward unified CNAPP architectures. IT service partners are no longer evaluated solely on their ability to deploy monitoring software; they must deliver integrated security platforms that correlate context across code, infrastructure, identity, and data.

Bridging the Gap Between Security Teams and Software Developers
Historically, cybersecurity departments and software engineering teams operated in friction-heavy silos. Security teams generated static audit reports, while developers viewed security checks as speed bumps that delayed release cycles.

By embedding Wiz’s contextual risk intelligence directly into EPAM’s software engineering workflows, the partnership enforces DevSecOps integration as an enterprise standard. Systems integrators and IT consulting firms must now deliver “security-by-design” frameworks ensuring cloud security controls are built natively into code repositories and deployment pipelines rather than applied as an afterthought.

Broad Operational Impact on Enterprise Businesses

For enterprise organizations navigating digital migrations while attempting to defend expanding attack surfaces, adopting an integrated cloud security model provides clear strategic and financial advantages:

Insulating Cloud Operating Margins Against Cyber Disruptions
Ransomware threats, cloud misconfiguration, and data breaches come at a significant cost in terms of fines, legal fees, and branding impact. The ability to consolidate cloud security posture helps organizations identify and eliminate vulnerabilities before they become exploitable by attackers, safeguarding their resources and maintaining their profit margins.

Reclaiming Capacity for Digital Innovation
With software developers spending valuable time investigating false positives and manually assessing security alerts, the process of digitization is slowed. The process of prioritizing risks and automatically remediating them in the workflow of developers frees up much-needed internal capacity. This enables the development teams to concentrate on delivering features that generate revenues.

The post EPAM Partners with Wiz to Accelerate Enterprise Cloud Risk Reduction and Cyber Resilience appeared first on ITDigest.

Also Read: Tech Mahindra and Cisco Partner to Deliver AI-Driven Security Service Edge Solutions 

Contextualizing the strategic importance of building technological autonomy directly into the underlying cloud architecture, Christian Müller, CEO of Schwarz Digits, explains: “Technological independence begins with infrastructure. Through this partnership, we are taking network security to a new level: we are evolving existing security architectures and focusing on high-performance identity verification. For our customers, this means even stronger protection against cyberattacks.” Reinforcing the vendor’s long-term enterprise commitment to data sovereignty across the region, Misha Kuperman, Chief Reliability Officer at Zscaler, added: “In an era of AI-driven cyber threats, a modern Zero Trust SASE solution is the only effective way to protect sensitive enterprise data. Our partnership with Schwarz Digits is a concrete demonstration of our long-term commitment to Europe: Together, we help our customers eliminate their attack surface and prevent the lateral movement of threats, while still maintaining compliance with European regulations through EU data residency. This enables the highest level of security while aligning with European digital sovereignty regulations.”

Read More: Zscaler and Schwarz Digits Partner to Deliver Sovereign Cloud Security Platform for Europe

The post Zscaler Partners with Schwarz Digits to Deliver Sovereign Cloud Security for Europe appeared first on ITDigest.