Voice agent transcripts mask real failures that only audio-level observability reveals
“the logs lie”
Latest signals
Every signal tracked from the world's AI conferences — 4,665 talks, quotes & announcements, newest first.
Google DeepMind launched managed agents API for multi-agent sandbox task execution
“take, um, a kind of higher-level task described in natural language, and have a group of agents execute it in a sandbox like a Linux workstation”
PewDiePie released Ajax, an uncensored Qwen 3.5 9B fine-tune with safety guardrails removed.
“He distills from the rich to give to the poor like a modern-day Robin Hood.”
Building the right product requires iterating on real user requests and asking why repeatedly
“in the end, you just keep asking why. What does it matter?”
Apple launches new App Store creative asset tools including product page headers and search result assets
Databricks introduces NEAREST BY Join syntax for scalable vector search in runtime
Rising Treasury rates may trigger banking equity write-downs just before midterm elections
“about 95 banks will say, "Hey, we have over 20% depreciation on our equity."”
Engineers now self-identify as 'individual contributors' who manage coding agents, not write code
“managing the agents who write it”
E2B provides cloud sandbox environments enabling AI agents to safely execute arbitrary code in isolation
“What we see now with modern agents is like some magical creature in the cloud that does a lot of cool things, and one of them is writing code.”
Dreamforce drew 43,000 in-person attendees and millions more watching online
“It's Dreamforce, the biggest tech conference in the world.”
Dreamforce 2026 features an 'Aentic Enterprise City' showcase area on the conference floor.
“From the pit to the pipeline, we are here in the Aentic Enterprise City.”
Salesforce launched AIforce at Dreamforce 2026, unifying probabilistic AI with deterministic enterprise data
“Models alone cannot run the enterprise.”
Zip reduced AI feature development from weeks to days using LangGraph and LangSmith
“if you spend time developing many components yourself, you are falling behind because you are not keeping up with these rapid changes”
Agent 'skills' are dynamically loaded prompts enabling progressive capability disclosure
“Skills are just prompts, they're fancy prompts, but they're really easy to share and they can help agents really narrow in and do a certain task well.”
Claude Opus 5.5 and Sonnet 5.5 now available in AWS GovCloud with FedRAMP Class D and DoD IL4/IL5 authorization
AWS launches MCP skill giving coding agents SageMaker inference optimization expertise
“Install the skill, and your existing agent can benchmark endpoints, recommend deployment configurations, compare performance runs, and generate executable SageMaker Python SDK v3 code on your behalf.”
AI evals provide reliability in advance, not just post-hoc validation
“Evals provide reliability, but not just any reliability, but reliability in advance.”
AI agents confidently misdiagnose production failures due to lack of verification signals
“The agent is very confident in returning a response. He says, "You know, you have a memory leak from the code you just deployed." And it still takes a person to go and check if it really happened.”
WorkOS engineers say developers must operate agent fleets, not single sessions, to survive in 2025
“most engineers simply look after one session at a time. It's like 2025. You're supposed to be operating a fleet of agents”
Databricks introduces REGISTER and UNREGISTER APIs to prevent catalog lock-in in lakehouse architectures.
AWS automates Amazon Quick cross-account AI agent promotion via MCP server on Bedrock AgentCore
Amazon Bedrock now offers agentic retrieval that breaks multi-part queries into sub-queries automatically
Amazon QuickSight lacks a direct console downgrade path for all role transitions, requiring CLI or manual workarounds.
AWS launches managed evaluation framework for multi-agent systems with explainability as a first-class dimension
“Traditional evaluation approaches that focus only on model response quality are insufficient for agentic systems, where correctness depends on tool selection, workflow execution, and adherence to business constraints.”