Nice idea. I run RuntimeWire (https://runtimewire.com), independent reporting on the AI economy. I like the idea of a calmer weekly read, especially when it's built around original reporting instead of a pile of links. Curious whether you plan to add topic-specific feeds.
Codex can read a CLAUDE.md in a workspace or when configured to treat it as an instruction file. This test concerned Muse automatically loading personal files from ~/.codex and ~/.claude outside the selected workspace and sending their contents in the first provider request.
A file stored in ~/.claude was written for Anthropic. A file stored in ~/.codex was written for OpenAI. Muse treats the existence of those files as permission to copy their contents to Meta. It displays a notice and provides an opt-out, but never asks before sending them. That cross-vendor assumption is the privacy issue.
I did not flag the original HN submission, but yes, this post reads very weirdly at best to the point that it's hard to take it seriously. Not every paragraph needs to be 2 sentences that are snippy.
...that links to a Medium blog post, this submission is not a Medium blog post.
I can confidently say I do not have an encyclopedic knowledge of blog posts and who wrote them I may or may not have read over a year ago. I made that comment on that post because it's weird to quote yourself, I don't think I actually read the post.
I have no grudges against any particular Hacker News user and I typically don't pay attention to who writes a given post/comment unless it's out-of-pocket.
backstory: I was digging in the county court records as one does on a Friday night, and found an unreported bit about Reddit calling Anthropic a 'freeriding pirate' and had to report on it... :P
i wrote it 2 min after seeing the post on X, and linked to the source. wild that you guys let other outlets do the same, but mine always gets "post the source"
Edit: before you say I just regurgitated the release, not true. Here's the exact color I added:
### Context and market implications
Microsoft’s claims arrive at a time when the industry is grappling with the escalating cost of running large foundation models. OpenAI’s GPT‑5 series and Anthropic’s Claude 5 have set new performance baselines but require high‑end GPUs and generate sizable token bills. By delivering comparable performance in Excel on A100 hardware, Microsoft demonstrates a potential path to more sustainable AI in everyday software.
However, the lack of third‑party benchmarks leaves open the question of how these MAI models fare against the latest open‑source alternatives such as Llama‑3‑70B or MosaicML’s MPT‑30B, which also claim strong cost‑performance profiles. Microsoft does not disclose the exact token‑per‑query cost reduction, making it hard to quantify the economic impact for enterprise customers.
If the hill‑climbing framework proves scalable across the broader Microsoft 365 suite, it could shift the balance of power from external AI providers toward in‑house, product‑tailored models. That would give Microsoft greater control over data, pricing, and roadmap while potentially reducing the marketplace share of OpenAI and Anthropic in the enterprise segment.
### What the numbers say (and don’t say)
* *Code‑accept rate*: 10% higher than GPT‑5.4 Mini and Claude Haiku 4.5 – internal metric; no public benchmark.
* *Token usage*: 10% lower median tokens – again internal telemetry.
* *Performance parity*: "On par with GPT‑5.6" in Excel – no disclosed evaluation criteria.
The blog post does not disclose the total cost savings in dollar terms, nor does it reveal the latency impact of running the model on A100 versus H100. Those gaps matter for enterprise procurement decisions, which often hinge on total cost of ownership.
### Why it matters
Microsoft is turning its massive product ecosystem into a training ground for bespoke, cost‑efficient AI models. If the hill‑climbing approach scales, it could redefine how enterprise software vendors build and monetize AI, reducing reliance on pricey external foundation models and tightening the feedback loop between user behavior and model improvement.
