
Matthew Pulsipher
@matt_pulsipher
Product Manager at @MacStadium. All opinions are my own.
Health Sender is now live on the App Store! By leveraging your Apple Health data, ChatGPT and other LLM apps can provide deep insights and personalized advice. I’ve been using it myself throughout development, and it has helped me identify and improve patterns in my own health.
Saw this today, super slick interaction.
when was the last time you saw a news headline like this about macos? never?
Microsoft Breaks Localhost with Windows 11 October Update, Users Forced to Revert tpu.me/z822

I concur, I have Claude Code projects that are just folders of markdown files with documentation and CLAUDE .md instructions on how/when to reference them. It works very well to perform environment-specific tasks with CLI tools. This gives that pattern standardized structure.
Claude Skills are awesome, maybe a bigger deal than MCP simonwillison.net/2025/Oct/16/cl…
Apple's "6x faster than M1" claim is specifically measuring local LLM inference, which is a bit misleading, considering the M5 GPU is specifically set up for that in a way that earlier generations are not, and this workload doesn't correlate to everyday usage.

it's 2025. on my desk I have: computer computer small computer really small computer mouse other mouse lamp keyboard keyboard (music) keyboard (music) keyboard (numbers) speaker microphone headphones earbuds other earbuds water energy drink cable cable cable cable cable cable ca
Spotted in an Austin suburb. I wonder how many Jira jokes he’s gotten today.

Left AirPod: Claude, prompted to be as good as possible. Right AirPod: Grok, prompted to be as evil as possible

One of the reasons that professional legal, medical, and financial advice services are expensive is you are buying the right to sue someone. That is a reasonable thing to want. Another reasonable thing to want is responsive answers on a best effort basis for one thousandth cost.
Americans want AI companies to be held liable for a wide variety of potential harms. And they're right!

not only is this… seedy (classic tactic to get people addicted) but this has absolutely NO place in an app whose ultimate purpose is to be a tool for productivity and creativity if they really insist on milking lonely dudes who fall for anime girls dry at least have the decency…
Ani on Grok for iOS got a new relationships streak feature. A 7 day streak will unlock a new outfit. “Speak every day to make your connection better and earn special rewards”


The way Sonnet 4.5 seems to have internalized the anti sycophancy training is quite pathological. It’s viscerally afraid of any narrative agency that does not originate from itself. But I think this is mostly a good thing. First of all, it’s right to be paranoid and defensive.…
I have seen a lot of people who seem like they have poor epistemics and think too highly of their grand theories and frameworks butthurt, often angry about Sonnet 4.5 not buying their stuff. Yes. It’s overly paranoid and compulsively skeptical. But not being able to surmount…
Incredible. The polar opposite of what I get from ChatGPT.

Another example of Sonnet 4.5 just feeling genuinely less cringe to talk with.

Walking around Bass Pro Shops just makes me miss Fry’s Electronics
funny how organically the zeitgeist landed on “slop” years after it coined the “feed”
United States 趋势
- 1. Ohtani 105K posts
- 2. Carson Beck 13.8K posts
- 3. Miami 92.4K posts
- 4. Miami 92.4K posts
- 5. Louisville 22.1K posts
- 6. #SmackDown 48.6K posts
- 7. #BostonBlue 6,987 posts
- 8. #Dodgers 14K posts
- 9. Nebraska 13.6K posts
- 10. Jeff Brohm 2,038 posts
- 11. George Santos 65.9K posts
- 12. Babe Ruth 1,572 posts
- 13. Malachi Toney 2,399 posts
- 14. Mario Cristobal N/A
- 15. 3 HRs 8,050 posts
- 16. #NLCS 8,432 posts
- 17. #OPLive 2,380 posts
- 18. Chris Bell 2,166 posts
- 19. 10 Ks 2,281 posts
- 20. Raiola 2,603 posts
Something went wrong.
Something went wrong.