AI Agents
Why AI Self-Reflection Fails Without External Feedback
Most people who work with AI long enough eventually discover the same trick. The first answer is not quite right, so instead of starting over, you send it back: Check your work. Find the weaknesses. Improve the answer. The second response often feels noticeably better. The argument is cleaner. An awkward paragraph disappears. The model […]
Claude Now Leads 26% of Anthropic R&D as AI Starts Helping Build AI
For most of the AI race, progress has been watched from the finish line. A new model arrives. Researchers measure how well it codes, reasons, solves mathematics or uses tools. The scores move upward, and the world concludes that AI has become more capable. Anthropic is now asking people to watch somewhere else: inside the […]
Delos Raises €10M as AI Workers Move From the Chat Window to the Org Chart
Most software enters a company through a login. Delos wants its software to enter through the org chart. The company describes the distinction directly: “Not a chatbot to prompt. Real colleagues with their own email, phone and digital computer. Proactive, autonomous and working 24/7.” Its AI Workers are given something closer to the setup of […]
Clay Raises $115M at $7.1B to Build Self-Learning Revenue Engine
Most sales software waits for a human to decide what should happen next. Clay is raising $115 million to see whether the software itself can become better at making that decision. The New York-based company has closed a Series D at a $7.1 billion valuation, led by Wellington Management, with backing from Sequoia, StepStone, a16z […]
OpenAI Says AI Agents Now Do 3.1x More Research Work Than Humans
At the start of 2026, coding agents were still operating in the background of OpenAI’s research organization, helping researchers write code, troubleshoot experiments and move technical work forward. Within months, that balance shifted dramatically. Before June, total agent runtime was still below human labor. By mid-August, OpenAI says researchers were consuming 3.1 agent-workdays for every […]
AIR Raises $50M to Secure AI Agents, Skills and MCP Servers
AIR’s researchers started with something that looked harmless: a useful AI skill. They placed it inside a respected open-source ecosystem, where it inherited the credibility of roughly 37,000 GitHub stars, promoted it to ordinary users, and watched security scanners approve it. People installed it. Eventually, AIR says the experiment reached more than 26,000 AI agents. […]
Apate.AI Raises $8.15M to Scale AI Scam Prevention Globally
Somewhere inside a scam operation, a fraudster can spend two hours persuading a victim to hand over money. For Apate.AI, the useful part is that the “victim” may never have existed. The Sydney-born company has built thousands of conversational AI agents designed to behave like believable scam targets. They answer calls and messages, keep fraudsters […]
Blue Machines AI Launches Project Icebreaker Ahead of GFF 2026: A Sovereign AI Co-Innovation Program for Indian BFSI
Blue Machines AI will partner with five most ambitious CX AI initiatives from Indian financial institutions and invest its multi-crore platform, proprietary technology, engineering and deployment capacity to take them into production Bengaluru, 31 August 2026: Blue Machines AI, an advanced agentic CX operating system for enterprises, today announced Project Icebreaker, a sovereign AI co-innovation program for […]
OpenAI Cuts GPT-5.6 Sol API Pricing as Frontier AI Enters a Cost Race
OpenAI is making its most capable GPT-5.6 model cheaper at a moment when developers are starting to judge frontier AI on more than intelligence alone. GPT-5.6 Sol will temporarily cost $4 per million input tokens and $20 per million output tokens, reducing the price of running complex reasoning, coding and agentic workloads through OpenAI’s flagship […]
Gemini 3.7 Flash Moves Complex AI Work Into a Lower-Cost Tier
The old model hierarchy was easy to understand. Bigger models handled the difficult reasoning; Flash models earned their place through speed and lower cost. Gemini 3.7 Flash is designed to make that division less obvious. Google is positioning its latest Flash release as a production model for coding, multi-step agents and knowledge work, while keeping […]
Mindgard Raises $30M for Attacker-Aligned AI Security
Mindgard began with the conclusion that traditional application security was looking at the wrong kind of problem. Research led by Dr. Peter Garraghan at Lancaster University had been examining AI security for more than a decadebefore Mindgard was founded in 2022. The challenge was increasingly clear: AI systems could behave probabilistically and opaquely, introducing exploitable […]
STRGY AI Raises €1M to Turn Corporate Strategy Into a Living System
A company can spend weeks defining its strategy and still lose sight of it once everyday work takes over. Priorities spread across teams, reporting arrives in cycles, and leadership may discover only weeks later that execution has drifted from the plan. Helsinki-based STRGY AI is building around that blind spot, turning strategy from a periodic […]