AI Journal 198th Edition

OpenAI Launches Frontier to Scale Enterprise AI Agents

OpenAI has launched Frontier, a new enterprise platform that unifies company data and systems to deploy AI agents at scale. Complementing ChatGPT Enterprise, Frontier helps businesses manage and optimize agents as AI co-workers, supporting OpenAI’s push to grow enterprise revenue toward 50%.

LET’S STEP INSIDE →

Anthropic AI Tools Rattle Software Stocks, Spark Long-Term Debate

New AI tools from Anthropic triggered a sharp sell-off in software and data stocks, raising fears of margin and pricing pressure. While executives argue the panic is overblown, analysts say AI may reshape valuations, with mission-critical software firms better positioned to coexist with AI.

LET’S STEP INSIDE →

Big Tech’s AI Spending Race Tests Investor Patience

Big Tech is pouring unprecedented sums into AI infrastructure, led by Amazon’s projected $200 billion capex through 2026, followed by Google, Microsoft, and Meta. While firms see compute as the key to AI dominance, investors remain uneasy about the massive costs and uncertain returns.

LET’S STEP INSIDE →

Microsoft Unveils Scanner to Detect Hidden Backdoors in AI Models

Microsoft researchers have introduced a new scanning method to detect “sleeper agent” backdoors in open-weight AI models. By exploiting memorisation leaks and attention patterns, the tool identifies poisoned models without knowing triggers, helping enterprises securely vet third-party LLMs before deployment.

LET’S STEP INSIDE →

Peak XV Faces Senior Exits Amid AI Push and U.S. Expansion

Peak XV Partners has seen senior departures following an internal disagreement, with three long-tenured partners leaving to launch a new VC firm. The exits come as Peak XV doubles down on AI investing, strengthens internal leadership, posts strong IPO gains, and prepares to open a U.S. office.

LET’S STEP INSIDE →

OpenAI has launched GPT-5.3-Codex, its most advanced agentic coding model yet, capable of handling complex, multi-step tasks while collaborating with developers and contributing to its own development.

LEARN MORE →

Around 35 of 85 countries at a military AI summit signed a non-binding declaration on responsible AI use in warfare, while major powers like the U.S. and China declined amid geopolitical tensions

LEARN MORE →

Operant AI has launched Agent Protector, a unified security platform for agentic AI, offering real-time visibility, threat detection, zero-trust enforcement, and governance to protect autonomous AI agents across enterprise environments.

LEARN MORE →

Ronnie Sheth, CEO of SENEN Group, stresses that enterprises must fix data quality before adopting AI, noting that strong data foundations are key to achieving practical, measurable value from enterprise AI initiatives

LEARN MORE →