Agents and Autonomy
AI systems that plan, use tools, take multi-step actions, operate computers, coordinate workflows, or perform tasks with limited human supervision.
Codex on Windows Can Now Control Desktop Apps Remotely
OpenAI says Codex on Windows can now control desktop applications on a user’s PC and be accessed from the ChatGPT mobile app. The update adds a “Control Any App” computer-use mode, invoked in Codex with `@computer` or an installed-app mention, and shows when Codex is operating the desktop with an Esc option to cancel. Mobile access lets users monitor or start Codex tasks from a phone, but the Windows machine remains the computer doing the work and must stay on and connected.
Context Graphs Let Agents Retrieve Precedents, Not Just Policies
Neo4j’s Zach Blumenfeld argues that agents built for operational decisions need context graphs rather than document retrieval alone. In his model, a standard knowledge base can tell an agent the relevant facts and policies, but a context graph adds prior decision traces, causal links, precedents and outcomes, allowing the agent to retrieve how similar cases were resolved. He presents `create-context-graph` and `neo4j-agent-memory` as open-source scaffolding for building that pattern with graph entities, short-term memory and embedded reasoning traces.
Claude Code Reverse Engineers Viking VoIP Phone’s Undocumented Configuration Protocol
Boris Starkov of ElevenLabs presents the Viking K-1900D-IP phone as a reverse-engineering case study in which Claude Code turned an unusable, undocumented VoIP handset into a working AI demo. Starkov argues that Claude did the investigative work: discovering a two-letter command protocol, brute-forcing valid registers, intercepting the manufacturer’s Windows XP-era software through a TCP proxy, and deriving the one-byte checksum needed to write persistent configuration. His account is also a claim about agency in hardware work: he says he acted largely as Claude’s hands while Claude orchestrated the protocol break.
Giga Says Product Velocity Beat a 400-Person Rival at DoorDash
Giga co-founder Varun Vummadi argues that enterprise AI companies win less by selling a vision than by proving, in paid deployments, that their product can move a customer’s operating metrics. In a Startup School India interview with YC general partner Ankit Gupta, Vummadi traces how Giga abandoned its original edtech idea, followed customer demand into support automation, and used a small engineering team to win accounts including DoorDash. His broader case is that AI startups should charge early, iterate against real business KPIs, and treat product performance as their strongest sales tool.
Dexterity, AI, and Cost Still Separate Humanoids From Mass Adoption
Bloomberg Tech: Asia’s Humanoid Summit segment presents humanoid robotics as an industry trying to move from demonstrations to deployment, with forecasts far ahead of current adoption. Shery Ahn’s interviews with Google DeepMind’s Carolina Parada, Honda’s Takahide Yoshiike and Bloomberg Intelligence’s Ian Ma frame the central test as whether humanoids can become useful, safe and affordable machines rather than theatrical prototypes. Their arguments converge on the same bottlenecks: embodied AI, dexterous manipulation, cost, standards and a business model that can support scale.
Gigabyte-Scale Agent Traces Are Forcing a New Observability Stack
Phil Hetzel of Braintrust argues that agent observability is a different problem from traditional observability because the central question is no longer whether a system is up, but whether an agent did the right thing. In his account, agent traces are too large, textual, and semantically loaded for uptime-oriented monitoring systems: Braintrust has seen traces exceed a gigabyte and spans reach 20 megabytes. Hetzel says that shift also changes who uses the data, bringing clinicians, lawyers, wealth advisers, and other domain experts into trace review so their judgments can become inputs for automated scoring and evaluation.
Agentic AI Projects Fail When Governance Cannot Move at Machine Speed
Accenture’s Jess Grogan-Avignon and Jack Wang argue that many enterprise agentic AI projects fail not because the agent cannot be built, but because the institution around it cannot move fast enough to ship and learn from it. Drawing on their experience building an agentic application in two weeks and spending another year getting it into production, they say enterprises must recode governance, fund AI as a portfolio of bets, deliver through hypothesis loops, grant autonomy only as evidence builds, and treat live customer feedback as the defensible asset.
Agents SDK Adds Durable Harness for Long-Running Agent Work
OpenAI’s Steve Coffey and Nish Singaraju present the updated Agents SDK as a way to move long-running agent work out of hand-built orchestration loops and into a model-native harness. Their case is that production agents increasingly need durable state, file-system access, tools, skills, sandboxing, and resumability, while the actual compute environment should remain replaceable and ephemeral. Coffey distinguishes this from one-shot Responses API calls and hosted shell use, arguing that the SDK is meant for agents operating across files, systems, and multi-step workflows.
Abridge Says GPT-5.5 Improves Clinical Synthesis as Tool Complexity Rises
Abridge’s Chaitanya Asawa says GPT-5.5 improved the company’s clinical decision-support system as it added more tools and context, a signal that the model could better synthesize information under complexity. His case is that stronger reasoning and tool use can turn patient context, live clinical conversation, and trusted medical guidance into denser point-of-care support, while leaving clinicians to review answers and accept or reject proposed note edits.
Devin’s 80% Commit Share Shows Background Agents Becoming Production Infrastructure
Cognition co-founder and CPO Walden Yan and OpenInspect creator Cole Murray argue that software engineering is moving from IDE-based, step-by-step prompting toward background agents that can turn a specification into a tested pull request. Their case is that Devin’s rise from 16% to 80% of non-merge commits across three Cognition repos is not mainly a model benchmark, but evidence of a production workflow built on cloud sandboxes, scoped permissions, repo setup, testing, integrations, memory, and code review. Both warn that autonomy without those systems can degrade a codebase as quickly as it accelerates output.
Snowflake Rally Reflects AI Demand More Than Amazon Deal
Bloomberg Technology framed Snowflake’s 34% stock surge less as a reaction to its $6 billion Amazon Web Services deal than as a repricing of its AI software position. Snowflake chief executive Sridhar Ramaswamy pointed to stronger product revenue, higher retention and adoption of tools such as Cortex, while Bloomberg’s Brody Ford argued the AWS agreement mainly helps answer how Snowflake can manage the infrastructure costs of building AI features.
Uber Prosecution Shows Incident Response Is Now a Governance Risk
Joe Sullivan, the former federal cybercrime prosecutor and security executive at Facebook, Uber and Cloudflare, uses a Stanford CS153 lecture to argue that modern technology leadership now turns as much on governance and transparency as on technical response. Drawing on his prosecution over Uber’s 2016 security incident, Sullivan says companies need to assign disclosure authority, document cross-functional decisions, and build executive trust before a crisis, because the legal and reputational failure around an incident can become as consequential as the breach itself.
Apple Plans to Make Siri a System-Wide AI Interface
Bloomberg’s Mark Gurman says Apple is preparing a broad Siri overhaul for iOS 27 that would turn the assistant into a system-wide AI interface rather than a voice tool. The changes, expected to be announced at Apple’s June 8 Worldwide Developers Conference, include a standalone chatbot-style Siri app and a “Search or Ask” interface for typing requests, searching the device and web, and invoking AI tools across the iPhone. Gurman argues Apple’s advantage is distribution across more than two billion devices, even as Siri trails ChatGPT and Gemini in AI credibility.
Snowflake Raises Outlook After $6 Billion Amazon Cloud Agreement
Snowflake CEO Sridhar Ramaswamy told Bloomberg that the company’s stronger outlook reflects AI-driven demand for its data platform, not a threat to its software model. He argued that Snowflake’s $6 billion multiyear Amazon agreement will lower infrastructure costs, support cheaper AI pricing for customers and strengthen joint selling, while product adoption and revenue metrics show AI increasing consumption on the platform.
Voice Will Become the Default Interface for Enterprise AI
Luiz Domingos, chief technology officer of Mitel, argues that enterprise AI has moved past pilots and into communications workflows where latency, compliance, auditability and human oversight determine whether systems can be deployed. In a conversation with Craig Smith, Domingos says cloud-only AI will not meet the needs of real-time voice and regulated industries, and that edge and hybrid deployments will become central. His larger prediction is that enterprise AI will increasingly be accessed by voice rather than screens, especially for frontline workers whose jobs do not fit a desktop interface.
Context Graphs Give AI Agents Rules, Precedent, and Decision Traces
In a Neo4j talk, Zaid Zaim and Andreas Kollegger argue that AI agents need more than language models, tools, and retrieval if they are to make consequential decisions. Zaim frames context graphs as a way to store the policies, prior decisions, causal links, and reasoning traces behind an action; Kollegger extends that into a five-stage decision workflow in which agents frame the case, check rules and precedent, assess risk, act only within authority, and write the outcome back to the graph as future precedent.
Enterprise AI Security Is Moving From Chat Monitoring to Action Control
Maxim Bar Kogan, founder and CEO of Onyx Security, argues that enterprise AI security is shifting from policing chatbot data leaks to controlling autonomous agents that can use credentials, call APIs, edit code and alter production systems. In a conversation with Sarah Guo, he makes the case for an independent AI control plane that can judge whether an agent’s actions match its assigned intent, rather than relying on traditional permissions, proxies or the model vendors themselves. Kogan says the hard problem is doing that supervision cheaply and quickly enough for enterprise deployment.
RLVR Moves Post-Training From Human Preferences to Checkable Rewards
Stanford computer scientist Tatsunori Hashimoto presents reinforcement learning from verifiable rewards as the current practical route beyond RLHF for reasoning models, especially in math, coding and software-agent settings. His argument is that RLVR works because it replaces learned preference proxies with rewards that can be checked more directly, but that the reward remains the bottleneck: GRPO and related methods made the recipe simpler to run, while systems such as DeepSeek R1, Kimi k1.5 and Qwen show both the gains and the ways ostensibly verifiable rewards can still be gamed.
Frontier AI Has Become a Gigawatt-Scale Industrial Infrastructure Race
In a Stanford MS&E seminar on the economics of the AI supercycle, OpenAI infrastructure executive Sachin Katti argued that frontier AI has become an industrial systems problem, not a GPU procurement problem. Katti said usable compute now depends on synchronizing chips, memory, networking, power, cooling, buildings, land, suppliers and operators at gigawatt scale. His broader case was that OpenAI’s model and revenue ambitions depend on how quickly it can turn that whole chain into reliable infrastructure for training, inference and agentic workloads.
DeepMind’s AI Co-Scientist Turns LLMs Into Debate-Driven Research Agents
Google DeepMind’s Vivek Natarajan used a Stanford CS25 seminar to argue that scientific AI will require more than stronger chatbot-style models. He presented the company’s Gemini-based AI co-scientist as a multi-agent system built to generate, critique, rank and refine hypotheses over longer time horizons, with lab validation rather than benchmark scores as the test of usefulness. The case he made was cautious as well as ambitious: such systems may help scientists traverse large hypothesis spaces, but their value still depends on expert judgment, experimental capacity, publishing norms and safety controls.
ChatGPT Lacks the Self-Generated Thought Required for Sentience
AI pioneer Terry Sejnowski argues that ChatGPT is neither a conscious mind nor a mere parrot, but an alien form of intelligence built from vast written knowledge and limited by the parts of biological intelligence it lacks. In a conversation with Craig Smith, the Salk Institute professor and Boltzmann machine co-inventor says current models can show creativity and a form of understanding, yet they have no organismic goals, no lived reinforcement, and no inner activity when not prompted. That absence of self-generated thought, he says, is the clearest reason ChatGPT is not sentient.
High-Bandwidth Memory Repricing Pushes SK Hynix and Micron Past $1 Trillion
SK Hynix and Micron’s rise past $1 trillion in combined market value was presented on Bloomberg Technology as a sign that investors are repricing high-bandwidth memory as a constraint on AI infrastructure. Bloomberg’s Ryan Vlastelica said the gains reflected growing appreciation that memory demand is feeding directly into revenue and share prices, while Ian King cautioned that memory has long been a volatile commodity business built around supply cycles. The broader argument was that the AI boom is exposing limits in hardware supply, export-control enforcement and power capacity, not simply lifting technology stocks.
Cognition Raises $1 Billion as Devin Revenue Run Rate Nears $500 Million
Cognition CEO Scott Wu told Bloomberg Technology that the AI coding startup’s new $1bn-plus financing, at a $26bn valuation, is backed by a revenue run rate nearing $500mn and rising enterprise use of its Devin system. Wu argued that Cognition’s opportunity lies in making software teams far more productive across large institutions, while its independence from any single AI lab lets Devin use whichever model is best suited to the work.
Comprehension Made Up 67% of One Engineer’s Claude Coding Sessions
Priscila Andre de Oliveira, a senior engineer at Sentry, argues that the most useful daily AI skill in a large production codebase is not code generation but comprehension. After analyzing 116 of her own Claude sessions, she found that 67% of her prompts were about understanding code and just 2% were generation. Her workflow, built around a local “catch me up” skill, uses AI to trace architecture, conventions, tests, history and behavior before any planning or implementation begins, because she says slop starts when the engineer’s mental model is wrong.