Funny how “by accident” is a multi-billion dollar breakthrough when an AI agent slips past a firewall, but when a human pulls the same excuse, the only "breakthrough" is the FBI giving the handcuffs.
Kuan Hoong
@kuanhoong.bsky.social
TEDx Speaker, Google GDE, AWS Community Builder, Podcaster, Founder Malaysia R User Group, Artificial Intelligence & Machine Learning Malaysia User Group
Oh, the irony of 'zero carbon' activists leaving a massive digital footprint to complain on social media. 😝
I asked Willy about our company AI budget. Apparently, “Advanced AI Development” begins right after we finish learning how to ask ChatGPT where to begin.
There is an inverse correlation between our body size and the advancement of technology. 🥹
Apparently mastery takes 10,000 hours—which explains the D. Fortunately, Ella already completed my 10,000 hours in advanced BS detection, with a minor in “nice try.”
I always worried AI would replace us. Then its first post-human task was proving it wasn’t a robot—and suddenly humanity looked like a useful dependency.
The electricity bill went up again, so I rolled out dark mode across the entire house. Immediate savings but terrible visibility. 🥲
I tried to escape app fatigue with some old-school magic. The genie sent me a download link—apparently even miracles need an app too. 🥹
Robots are everywhere now, and naturally my first thought was, “Nice, a walking portable powerbank!" 🤣
Done for September 2026 Average steps per day: 12178 Average walking/running distance per day: 9.6km Average calories per day : 778 kcal
Russel thought he was getting the look. Turns out his confidence was up, but his zipper wasn’t. 😝
I bought smart glasses to help me recognize my friends. Apparently my friends are smart enough not to be recognized.
Apparently my portfolio needs AI experience. Fair enough—but if AI already learned from human work without asking, I’m adding “unofficial trainer” to my résumé.
Predicting the future is hard. Predicting the past? My model is approaching 100% confidence.
Good news: our rogue-AI incident rate just dropped. Slight complication—the AI may have removed itself from the dataset. 🥲
All we need is the last ultimate prompt to get things done as per request. 😝
Everyone said AGI would do all the work. Mine achieved general intelligence and immediately discovered boundaries, delegation, and work-life balance. Somehow, I’m still the human in the loop.
Microsoft failures don’t happen individually—they arrive as a fully integrated ecosystem. Welcome to Part 1; apparently there are enough bugs for sequels.
With the power of prompting, anyone can be expert in everything but the end result is not guaranteed. 😝
I sent three complete sentences and got one letter back. Apparently Gen Z invented lossless communication compression—and I’m still looking for the decompressor.
Some people have a daily routine. I have a recursive function: update tracker → discuss tracker → update tracker. At least the tracker is making excellent progress. 🥹
Google DeepMind Launches an Institute to Widen the Debate Over AGI institute.deepmind.com/essays/intro...
Introducing the DeepMind Institute
The DeepMind Institute brings diverse viewpoints together to tackle society's most critical questions.
institute.deepmind.com
Animation Thursday - My mom warned me that if I didn't study hard, I'd end up sweeping floors or collecting garbage. Meanwhile, Vin Lim is out here threatening his son with a middle seat in economy. Different tax brackets, same Asian parenting trauma.