Near-term agents get more tools.
Parts 1–9 mapped chat → reasoning → tools → loop → risk → fit → policy → eval → ship. This last part names patterns already peeking through. None of them require science-fiction timelines. All of them still need the same recommendation: goals, tools, checkpoints.
Multi-agent systems
Instead of one generalist loop, specialists pass work to each other: researcher → writer → reviewer. The mechanism is still tool calling plus a router. The new failure mode is a bad handoff: the reviewer never sees the tool error the researcher swallowed. Eval the chain, not only each persona’s last paragraph.
Agentic RAG
Retrieval-augmented generation (external link) (Lewis et al., 2020) was “fetch documents, then write.” Agentic RAG is retrieve from your approved knowledge, then act only within those limits: open the ticket the doc describes, fill the form the policy allows. To safeguard data, retrieval needs ACLs (Access Control Lists). Retrieval plus tools is an agent with a library card, which is better than an agent with the whole internet, if the corpus is actually yours and actually current.
Computer-use agents
APIs are the sane default. Computer-use is the escape hatch when there is no API: the model looks at screenshots and returns clicks and keystrokes. OpenAI’s computer-use API guide (external link) describes that loop: screenshot → action → new screenshot. Anthropic’s computer use tool (external link) describes the same pattern from another company. Permissions get harder, not easier. A click is a write to whatever is on screen. Human confirmation belongs on anything that looks like purchase, send, or admin.
AI-native applications
Products built around a goal and a toolbelt, not a chat box bolted onto last year’s UI. The user states an outcome. The app runs a constrained agent. Chat is a log, not the product.
Autonomous software development
Longer coding loops where tests decide success: agents that open branches, run CI, and ask for review. Judgement, product risk, and “should we build this?” stay with people. Autonomy here means fewer keystrokes, not fewer owners.
Some products worth mentioned
The patterns above already have product instances.
OpenClaw (external link) is a self-hosted gateway: one process that holds sessions, tools, memory, and channel plugins: chat apps in, agent out. That is an AI-native app with a toolbelt you run yourself, not a new layer in the stack.
Claude’s computer-use tool (external link) is Anthropic’s version of screenshot → click → new screenshot. OpenAI’s agents guide (external link) is one SDK/runtime version of a constrained loop. Neither replaces the gates from Parts 7–9.
What not to wait for
A single agent that “does the company.” A guarantee that computer-use is safe because the demo was smooth. A future where evals become optional.
The basic structure from Parts 7–9 does not expire when the pattern names change.
Conclusion
We made AI good at answers. Then we let it think longer. Then we gave it tools. The next phase is more of that loop (more agents, more retrieval, more UIs), only if a person still approves the steps that cannot be undone.
If you read one habit into this quarter, make it this: stop asking only “What should I prompt?” Start asking “What may it touch, what must I approve, and how will I see what it did?”
Takeaway: What’s next is still the agent loop (multi-agent, retrieve-then-act, computer-use, named products included) with the same gates you already designed.