> Markdown version of [/videos/1930-a-true-story-about-speeding-up-the-wrong-things?t=580](https://www.wearedevelopers.com/videos/1930-a-true-story-about-speeding-up-the-wrong-things?t=580). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# A True Story About Speeding Up the Wrong Things

Are you shipping code faster, or just accelerating your mistakes? Discover why surviving AI-generated complexity means prioritizing explicit steering systems and operational memory over raw output speed.

- **Speakers:** [Jo Hasenau](https://www.wearedevelopers.com/@jo-hasenau), [Petra Hasenau](https://www.wearedevelopers.com/@petra-hasenau)
- **Event:** World Congress 2026 Europe - Virtual Stage
- **Published:** July 1, 2026
- **Duration:** 39:40
- **URL:** https://www.wearedevelopers.com/videos/1930-a-true-story-about-speeding-up-the-wrong-things

## Summary

The tech industry is trapped in a cyclical obsession with acceleration, consistently prioritizing speed, scale, and output over feedback, responsibility, and control. From the dot-com bubble to the 2010 flash crash, the failure mode remains the same: systems grow faster than our ability to steer them. Artificial intelligence is not the origin of this problem, but it is the fastest and most powerful amplifier yet. As AI drives the cost of generating code and content toward zero, the engineering bottleneck shifts from producing artifacts to understanding which artifacts should actually exist. When systems depend on AI-assisted tools without adequate governance, they suffer from "fluent uncertainty"—producing highly plausible but unverified output, masking the reasoning behind decisions, and creating a dangerous responsibility blur where no one can definitively explain why an architecture behaves the way it does.

To survive the explosion of AI-generated complexity, organizations must look backward to cybernetics and Ashby's Law, which states that a control system must possess as much internal variety as the environment it regulates. Simply adding more autonomous agents or point-to-point API tools only multiplies unchecked relationships and results in "computational déjà vu," where machines expensively repeat the same mistakes without learning. Instead, engineering teams must build explicit steering systems and a clear "commitment layer" where AI drafts are rigorously validated into accountable system states before they become operational reality. Because AI chats often swallow the trial-and-error context that once lived in public domains like Stack Overflow, capturing the entire path of reasoning—not just the final patch—is critical to preserving broader operational memory.

The ultimate goal requires building a federated memory layer, or "mesh of knowledge nodes," where integrated systems can share verified patterns and failed assumptions without centralizing every truth or generating restrictive "governance theater." Crucially, the primitives of this control layer must remain open-source to ensure genuine technological sovereignty rather than managed dependency. Moving forward, the most successful engineering teams will not be those that merely generate the most AI output, but those that treat organizational memory, continuous feedback, provenance, and structured commitment as first-class architectural artifacts.

**Keywords:** ai governance theater, ashby's law, system steerability, fluent uncertainty, output inflation, agentic brute force, engineering responsibility blur, federated memory layer, architectural control variety, computational deja vu, knowledge mesh, open source steering primitives, software commitment layer, llm plausibility, generative ai infrastructure

## Chapters

1. **The historical pattern of confusing speed with progress** (00:00) — Unchecked technological acceleration frequently outpaces our ability to maintain feedback, responsibility, and control within dynamic systems.
1. **Questioning the rush for speed over strategic direction** (03:47) — Accelerating local metrics without understanding broader systemic optimization goals often generates more chaos than value.
1. **Learning from historical instances of unchecked market acceleration** (06:45) — Previous technology bubbles reveal that expanding reach and scale only amplifies underlying structural weaknesses.
1. **Abstracting complexity and the illusion of distributed responsibility** (09:40) — Financial markets and software architectures wrap risk in elegant abstraction layers that ultimately hide unmanageable exposure.
1. **The widening gap between automated action and systemic control** (11:00) — Executing tasks at machine speed turns crucial human oversight into retrospective archeology instead of active steering.
1. **Applying cybernetics and the law of requisite variety** (13:47) — Internal system variety must continually increase to successfully absorb the rising complexity of external operating environments.
1. **Identifying plausibility and output inflation in AI workflows** (18:04) — Treating plausible generation as absolute truth artificially deflates coding costs while drastically increasing the burden of judgment.
1. **The risks of replacing understanding with agentic brute force** (20:59) — Relying on automated trial and error loops bypasses the deep reasoning required for sustainable software engineering.
1. **Connection explosion and the multiplying complexity of direct integrations** (22:16) — Letting every automated tool and agent interface directly generates an impossible web of undocumented dependencies.
1. **Vanishing shared learning in private AI coding assistants** (23:08) — Moving development troubleshooting into private chat windows destroys the observable paths and failures that teams learn from.
1. **Navigating responsibility blur in automated software delivery pipelines** (24:06) — Inserting generative agents into agile processes diffuses accountability until nobody understands the system's final shape.
1. **Moving past governance theater to actively steer AI** (25:01) — Superficial compliance documents completely fail to constrain computational models that execute directly against strict production deadlines.
1. **Designing steerable architectures with explicit commitment layers** (27:14) — Isolating suggestion generation from verified state changes ensures models remain bounded by strict operational boundaries.
1. **Building a federated memory layer to preserve engineering knowledge** (29:53) — Retaining the rejected attempts of agentic workflows saves computational energy and strictly prevents repeating known mistakes.
1. **Governed service layers and open source control mechanisms** (34:42) — Centralizing routing policies and demanding open control primitives reduces the absolute chaos of uncontrolled model agent interactions.
1. **Practical steps for treating reasoning as an engineering artifact** (37:25) — Incrementally auditing single workflows ensures your engineering culture actively preserves vital feedback and operational memory.

## Related Moments

- [Navigating developer bottlenecks and human accountability](https://www.wearedevelopers.com/videos/100265-fireside-chat-in-conversation-with-werner-vogels-cto-of-amazon-com) (from "Fireside Chat - In conversation with Werner Vogels, CTO of Amazon.com")
- [Redefining the software architect role for AI pipelines](https://www.wearedevelopers.com/videos/100190-architecture-3-0-from-90-to-99-999-reliability-in-building-ai-systems) (from "Architecture 3.0: From 90% to 99.999% Reliability in Building AI Systems")
- [Pausing hyper-speed AI deployment for responsible leadership](https://www.wearedevelopers.com/videos/100213-the-human-api-designing-organizations-for-judgment-not-just-execution) (from "The Human API: Designing Organizations for Judgment, Not Just Execution")
- [Addressing institutional inertia and AI pilot failures](https://www.wearedevelopers.com/videos/100253-ai-in-high-stakes-industries-lessons-learned) (from "AI in High-Stakes Industries: Lessons Learned")
- [Aligning artificial intelligence safeguards with corporate realities](https://www.wearedevelopers.com/videos/909-edit-your-future-queerverse-radical-ai) (from "Edit Your Future: Queerverse Radical AI")
- [Rethinking software engineering processes beyond human constraints](https://www.wearedevelopers.com/videos/1830-wearedevelopers-live-speculaitions) (from "WeAreDevelopers LIVE - SpeculAItions")

## Related Articles

- [Never delegate the understanding](https://www.wearedevelopers.com/magazine/749-never-delegate-the-understanding)
- [WWC24 Talk - Scott Hanselman - AI: Superhero or Supervillain?](https://www.wearedevelopers.com/magazine/469-wwc24-talk-scott-hanselman-ai-superhero-or-supervillain)
- [What is Software Engineering in the Age of AI?](https://www.wearedevelopers.com/magazine/640-what-is-software-engineering-in-the-age-of-ai)
- [Exploring AI: Opportunities and Risks for Developers](https://www.wearedevelopers.com/magazine/522-exploring-ai-opportunities-and-risks-for-developers)

## Related Jobs

- [Principal Software Engineer, Enterprise AI Platform](https://www.wearedevelopers.com/jobs/ext/1467292-principal-software-engineer-enterprise-ai-platform) at **GitHub**
- [Staff Software Engineer, Copilot Experiences](https://www.wearedevelopers.com/jobs/ext/164361-staff-software-engineer-copilot-experiences) at **GitHub**
- [Senior AI Agent Software Engineer (Go, Python) (m/f/x)](https://www.wearedevelopers.com/jobs/48277-senior-ai-agent-software-engineer-go-python-m-f-x) at **Dynatrace**
- [AI Operations Manager (all genders)](https://www.wearedevelopers.com/jobs/48263-ai-operations-manager-all-genders) at **envelio**
- [Head of AI Applications](https://www.wearedevelopers.com/jobs/ext/1456210-head-of-ai-applications) at **ZEISS Group**
- [Head of AI Applications](https://www.wearedevelopers.com/jobs/ext/1231536-head-of-ai-applications) at **ZEISS Group**