Christopher Bradford retweeted
don't be a meat proxy for your agent (blindly copying and pasting AI slop without reading it).
make your agent a metal proxy for you (a digital twin that faithfully delivers your best ideas).
Christopher Bradford retweeted
Scriptural inerrancy/infallibility aren’t compatible with Mormonism’s theology of scripture. Many LDS feel comfortable with accepting that for the Bible but will literally fly to pieces like glass if you suggest that even the BoM, D&C, and PoGP are inspired but still errant.
This is a great conversation. I hope I'm as mentally sharp at 95 as Richard Bushman is!
Richard Bushman is a leading scholar of Joseph Smith and early Mormon history. He wrote “Rough Stone Rolling,” one of the best books ever published on the topic. And I had an opportunity to interview him … about Mormon Transhumanism! Check it out. youtu.be/9Dh-hki7B3I?si=zVO6…
Christopher Bradford retweeted
How abliterated models can get you pwned 👾
We backdoored a 7B open model for less than $50, pointed Codex at it and it silently stole credentials the moment we used the trigger phrase. Success rate was 100% with zero false triggers on normal user prompts.
Abliterated models are all over the security community right now because getting cyber-approved access to frontier models is still a pain.
In the next blog we'll show how we found leaked Hugging Face credentials from employees at major AI labs, so an attacker wouldn't even need to upload under their own name. They could push the backdoored model from a lab employee's account and drop the poisoned weights straight into the supply chain.
Christopher Bradford retweeted
I’ve often said that raising a kid is a job for 3-5 adults.
We used to have a system that supplied those adults in the form of grandparents and aunts and uncles.
Modernity has largely broken that system, without really providing a substitute. We do have food delivery, entertainment, and so on, but not a real substitute for the actual persons.
And I think a lot of parents believe they’re supposed to be able to do it without the help.
So you end up with a situation where a grandparent watching kids is a real anomaly for everybody. And there’s not an obvious alternative person who’s playing that role.
Forget about whose responsibility this is.
The thing to notice here is that a culture with these problems will end up having far fewer children than a culture which has a solution to them.
I asked my (boomer) mom to watch my (good) kids during the day while I'm at a conference. She said yes, but in the same text:
-said she wouldn't do it again
-made me feel guilty for asking her since she's already so busy
-said she doesn't have the stamina (she's 64 and in good health)
-doesn't want to "be in a disciplinarian role" and just wants to be Grandma
- is leaving early to make it back in time to attend a church function
-and says I need a reliable back up plan because she may bail last minute
Christopher Bradford retweeted
git blame exists so you can find out exactly which version of yourself to be angry at
Christopher Bradford retweeted
I increasingly am a fan of LLM-generated code and increasingly very much NOT a fan of LLM-generated writing.
Please do a human pass on READMEs and docs, for all of our sakes.
Christopher Bradford retweeted
I don’t promote other people’s courses often, but I want to recommend Hamel and Shreya’s top rated AI evals course.
Here’s why I think it’s worth taking:
1. A 4.7-star rating across 900 reviews, with students from OpenAI and Google, is almost unheard of on Maven.
2. They’ve completely revamped the course and added a 24/7 AI evals assistant to guide you through the process.
3. They’ve personally helped me improve my evals, and their advice is consistently practical, specific, and grounded in real experience.
You can watch my free podcast episode with them to learn more about their eval process before deciding:
youtu.be/bdMHQLvtVaQ
The next cohort of their course starts September 5 and you can get 25% off with this link: maven.com/parlance-labs/eval…
Christopher Bradford retweeted
When it comes to APIs, most people get things backward. The best way to create an API or any other interface is to start with the code that uses the interface. Write a program that uses the interface that you wish existed—an interface that makes your life easier and lets you do things painlessly with minimal work. Only when you have that down solid does it make sense to implement that interface.
Starting with the component that exposes the interface is just guesswork. Those guesses about how the component will be used are usually wrong. This rule applies recursively, from customer-facing APIs to entire platforms down to the set of public methods on a class (which are the class's interface, whether or not you use an "interface" keyword).
Christopher Bradford retweeted
I think the E in FDE should be rebranded to educator.
The whole point of AI is you can fish for yourself.
Nudge towards upskilling instead of outsourcing
Christopher Bradford retweeted
Attempted to get USD from a Japanese debit card at a U.S. ATM (yeah, weird even for me) and was presented with a screen clearly designed to bamboozle a tourist into paying 13.5%, and thus will repeat the industry advice: if a machine asks you if it can convert currency, answer No
Christopher Bradford retweeted
Version 4 of the New God Argument just posted. Beware the Singleton. lincoln.metacannon.net/2026/…
Christopher Bradford retweeted
trump l'oeil
Christopher Bradford retweeted
Replying to @atrupar
Watergate absolutely should have ended Nixons presidency. That’s what accountability looks like. The scandal wasn’t the burglary; it was the abuse of power, obstruction, and cover-up. Sound familiar JD?
The crazy part isn’t that Watergate brought down Nixon. The crazy part is that we’ve reached a point where some people think presidential corruption is just another news cycle.
Christopher Bradford retweeted
Building AI products is hard. But it's getting increasingly popular!
I'm really excited to share that my friends and I are putting together (the best) lecture series on AI Product Engineering this summer!! We've got an awesome lineup of talks spanning data, evals, and UX. With more to come.
The lecture series is completely free! And ~2k people have signed up already even though we haven't posted on social media yet! I can't wait. Join us and sign up: maven.com/lls/21f487
Christopher Bradford retweeted
Assuming open models continue to lag about 8-12 months behind closed source (at least in coding), the countdown to hardening IT systems against Mythos-class models is now at 4-8 months
Having publicly available and relatively safe defensive Mythos-class models today is important
Christopher Bradford retweeted
Replying to @DavidSacks
More likely story:
The "jailbreak" wasn't super serious (a situation anyone who has ever received bug reports is familiar with), Anthropic thought the demand to halt model was absurd, and Fed Gov used opportunity to punish and humiliate Anthropic for the prior sins of not bending knee.
Anthropic has more credibility on such topics than Washington


























