OpenAI introduces real-time voice models for smarter AI conversations
OpenAI announced a new generation of real-time voice models through its API platform, aiming to make AI voice interactions more natural, intelligent, and action-oriented. The release includes GPT-Realtime-2 for advanced conversational reasoning, GPT-Realtime-Translate for live multilingual translation, and GPT-Realtime-Whisper for real-time speech transcription. These models are designed to support use cases such as AI voice assistants, customer service automation, live captioning, and multilingual communication. OpenAI highlighted that the new models can maintain conversational context, handle interruptions, and even perform tool-based actions during voice interactions.
 
Gartner predicts AI will reshape data and analytics leadership by 2030
Gartner released its top predictions for Data and Analytics (D&A) in 2026, highlighting how AI will fundamentally reshape enterprise operations, governance, workforce strategy, and productivity tools over the next several years. Gartner predicts that by 2027, 75% of hiring processes will assess AI proficiency, while AI agents and generative AI are expected to disrupt mainstream productivity software markets with a projected $58 billion market shift. The report also forecasts that autonomous AI agents will increasingly manage governance policies, compliance enforcement, and data contracts by 2030. Gartner emphasized that organizations succeeding in the AI era will combine strong AI capabilities with human relational skills, governance frameworks, and semantic data infrastructure.
 
NetSuite 2026.1 expands SuiteCloud with AI-native development and advanced REST APIs
Oracle NetSuite announced major enhancements to the SuiteCloud Platform in NetSuite 2026 Release 1, focusing on AI-native development, expanded REST APIs, and next-generation extensibility. Key updates include the new AI Connector Service, which enables secure integration with AI platforms such as ChatGPT, Claude, and MCP-compatible systems through natural language interactions. NetSuite also introduced the SuiteCloud Developer Assistant, allowing developers to generate and refactor SuiteScript 2.1 code using AI prompts. In addition, 2026.1 closes critical REST API gaps to accelerate the transition away from legacy SOAP integrations, while new extensibility capabilities improve SuiteApp lifecycle management and AI-driven automation.
 
SAP advances toward an autonomous supply chain with AI-driven automation
SAP unveiled new advancements toward a more autonomous supply chain at SAP Sapphire 2026, focusing on embedding AI directly into planning, logistics, manufacturing, and asset operations. Central to the strategy is the expansion of Joule AI assistants and autonomous agents capable of orchestrating workflows across supply chain systems with real-time business context. Key innovations include enhanced SAP Integrated Business Planning (IBP), automated transportation load building, vendor-managed inventory capabilities, and AI-powered exception handling to reduce operational delays and inventory inefficiencies. SAP also introduced tighter integration between business data, AI governance, and enterprise workflows through its Autonomous Suite.
 
Thousands of “vibe coding” apps leak sensitive data across the open web
A new investigation revealed that thousands of rapidly created “vibe coding” applications, apps that are built largely through AI-assisted coding tools with minimal security review, are exposing sensitive corporate and personal data online. Researchers discovered publicly accessible databases, API keys, authentication tokens, customer records, and internal business information embedded within these apps. Many developers relied heavily on generative AI coding assistants to accelerate development but failed to implement proper security practices such as credential protection, access controls, and code validation. Organizations face increased risks of data breaches, unauthorized access, compliance violations, and reputational damage as insecure AI-generated applications rapidly proliferate across the web.
 
Hackers launch a 5-hour “low-and-slow” DDoS attack to evade detection
DataDome revealed details of a highly sophisticated “low-and-slow” DDoS attack that generated 2.45 billion malicious requests within just five hours against a major content platform. Unlike traditional volumetric attacks, the attackers distributed traffic across 1.2 million IP addresses and over 16,000 autonomous systems, allowing each source to remain below conventional rate-limit thresholds. The attack peaked at more than 205,000 requests per second while avoiding detection by standard defenses. DataDome successfully mitigated the attack in real time using multi-layered behavioral analysis, server-side fingerprinting, and threat intelligence without disrupting legitimate users. The incident highlights a major shift in cyberattack strategy. From brute-force flooding to stealthy distributed evasion and underscores the growing need for AI-driven behavioral security models capable of detecting coordinated malicious activity at internet scale.