Brian Klaas

don’t try to be the smartest in the room, be the one learning the most.

An error with applying tech, AI or otherwise, to education is to assume that students want to learn most topics voluntarily, and if given the right tools, will self-teach. But the truth is that intrinsic motivation is usually not enough, especially because learning is often hard

Wow, this "What We Learned from a Year of Building with LLMs" post is excellent. The authors give very specific advice about prompting techniques, information retrieval, evaluating workflows, monitoring, and more. oreilly.com/radar/what-we-…

@gpickin @GoIntoTheBox Thank you, sir! 😁

@brian_klaas - re-reading your entire blog on the plane on the way to @GoIntoTheBox - and had to just stop and say thank you! it is an absolute goldmine of #aws #lambda #stepfunctions and more. Thanks so much for sharing, caring, and presenting as much as you do sir! #cfml

🚨Our new paper on how instructors can use AI to create research-based experiences that would have been impossible to do before the advent of generative AI. We also teach how to build them. Blog on democratizing innovation: oneusefulthing.org/p/innovation-t… Paper: papers.ssrn.com/sol3/papers.cf…

Without legacy media, people just post unclear charts without data sources to make sweeping arguments with no risk of accountability or repercussions.

Calling all ColdFusion enthusiasts! If you have unique expertise or stories in #ColdFusion, apply to speak at the #CFSummit2024 in Las Vegas. Share your insights and inspire a global audience. Don’t miss the deadline: applications close May 31. Apply now: bit.ly/3Q5Jb6A

Posted by @carehart: "Recordings available for the recent 17-session Adobe ColdFusion Summit Online 2024" bit.ly/49pWVjD #coldfusion

This plot presents a sad truth: Media has failed it's duty to inform the public Instead, a business model of selling fear, stoking conflict, pandering, and pushing advertisers propaganda. McKinsey-ification and it's consequences has been a disaster for the human species 🧵

Did the Simpsons really do it again, predicting something like the Apple Vision Pro?

I’m gonna give 10 random people that repost this and follow me $25,000 for fun (the $250,000 my X video made) I’ll pick the winners in 72 hours

@emollick We’re offering this kind of course at Johns Hopkins Public Health. Covers how LLMs work, better prompting, differences between major tools/models, how it can support you in research, and a lot on the ethical/harm issues surrounding current generative AI. courseplus.jhu.edu/core/index.cfm…

Universities should consider putting together a single class on generative AI that covers the hot button issues: LLM basics, AI literacy, ethics issues, how to prompt & use AI, etc. Otherwise a lot of this content is partially taught & is ad-hoc scattered across various classes.

Folks need to stop publishing working papers saying "We tested it and AI can't do X" when: 1) The AI in question is GPT-3.5. It is obsolete and not telling us anything about capabilities. 2) There is no attempt to do any prompt engineering. Better prompts can solve many problems.

This remains a better understanding of the actual meaning of free speech than basically any of the dominant positions today and all of the "thick free speech" vs "thin free speech"/"free speech culture" vs "cancel culture" muddiness has confused your thinking.

Populism reduces GDP per capita by 10 percent over 15 years. aeaweb.org/articles?id=10…

🚀 Exciting news for AWS Step Functions users! Introducing "Redrive" for Standard Workflows – restart failed executions from point of failure, boost efficiency, and cut costs. Learn more: aws.amazon.com/blogs/compute/… #AWS #Serverless #WorkflowAutomation #preInvent

People with a victim mentality are more likely to engage in everyday vigilantism: that is, to monitor the people around them for moral infractions, and punish those who step out of line. psypost.org/2023/11/having…

Really interesting perspective - AI is going to be so ubiquitous that it's not going to help businesses differentiate themselves because everyone has it, and it's gonna end up being a tax that everyone has to pay to a few select companies. 🤯 youtu.be/Y3AfplTpX3Y?t=…

"It was surprisingly hard to teach an AI model to be antisemitic enough," he continued.

“Beauty is truth, truth beauty,—that is all Ye know on earth, and all ye need to know.” Keats was right about our instincts. This paper shows we view more beautiful infographics as more trustworthy, and they get more “likes” & more scientific citations. osf.io/preprints/psya…