About the Project
Why I built Hawes Insights
I built Hawes Insights to explore how AI could help research manufacturing news and turn it into useful briefings. This site shows the workflow and its original output. New briefings are currently paused.
The articles retain the writing style developed for the project.
Who the briefings were written for
Business leaders
Research on automation investment, workforce changes and industrial policy for people setting business priorities.
Plant and operations teams
Briefings on robotics, operational technology security, digital twins and supply chains for plant and operations teams.
Technology teams
Analysis of AI readiness, industrial connected devices, edge computing and the work needed to put these systems into practice.
Investment and strategy teams
Research on manufacturing technology, industrial policy and related risks.
How the workflow worked
From news feeds to published articles.
1. Finding news
RSS feeds supplied manufacturing, automation, workforce, cybersecurity and industrial policy news. The collection stage removed duplicate items for later review.
2. Assessing relevance
News items were scored for relevance, impact, timeliness, novelty and actionability. Higher-scoring candidates advanced to research.
3. Research and writing
The research stage looked for supporting sources, conflicting evidence, industry benchmarks and regulatory context.
4. Editorial review
The workflow included an 18-point checklist covering factual integrity, source verification, analysis and readability.
Alan Hawes
My background spans technology, service delivery and organisational change, including roles at Telstra and nbn. I’m based in Melbourne, Australia.
- Background
- Technology, service delivery and organisational change
- Approach
- AI-Assisted Research and Writing
- Focus
- Research and editorial review
The editorial approach
The project aimed to produce clear manufacturing analysis supported by named sources. Its editorial checks were part of the process, not a guarantee that every claim was correct.
Sources and evidence
AI assisted research and drafting. The editorial approach required factual claims to be traced to named sources.
Choosing what to cover
The scoring process selected relevant news for research and manufacturing analysis.
Reading the archive
The original workflow used multiple AI agents to research topics, check sources and draft articles in an editorial voice developed for the project. The articles remain here as examples of that output, with their original wording, publication dates and source links.
AI-assisted research and writing can contain errors. Read the analysis and forecasts in the context of their publication dates. The archive has not been fully rechecked against current information.
The illustrations were added for this showcase. They are AI-generated concepts, not photographs of the companies, products or events described.
Some linked sources may have moved or become unavailable. A broken link alone does not establish that a claim is wrong.