I just want to say. I recently came across your name. Over the whole wannacry situation. I just want to say thank you. You saved many lives that would have been lost in hospitals. You are a hero. Also for what you do now
This is off subject but just watched your interview with David B from last year. You had mentioned that your content isn't long enough for RU-vid but I for one appreciate terse videos that get to the point no matter the platform. Keep uploading when you can!
from a kid who made a banking trojan (unwillingly) to a teen who saved the millions from being robbed, to a man that teaches you how to improve yourself.... Good Job man, Good Job. @MarcusHutchins (also this is the first time in my 35 years on this planet to tell someone they did a Good Job)
Absolutely will be using ChatGPT to learn from all my code that doesn't work (so much in c ...doesn't work) and holy cow- "Losing Your Will To Live"...hilarious!
The first code explanation I turned to was the Denzel bar "Explain This To Me Like I'm a Two-Year-Old." I noticed you also featured that; excellent video!
I only read two programming books in my entire career in embedded programming: K&R C programming 2nd ed, and a text on Real Time Programming. With today's complex environments and IDE toolsets, I do not envy anyone starting out trying to master the infinite complexity. Looking back, I wished I had taken courses in RTOS, Git, XML, Coding standards, and TCPIP.
This clicked for me. I can apply this same methodology to learning infosec and cybersecurity. Like you said use books for reference. and ask chat gpt to explain things a bit better.
So I worked in corporate tax accounting prior to pivoting to a more data driven field, and I tried quizzing Chat GPT on aspects of the US IRC. The info it provided was around 9 years out of date.
There are many models (some older than others) and some models rely on different sources to investigate and find their information. Some of those sources can be out of date, thus handing you late information. Just like browsing the internet, I wouldn't entirely rely on 1 source (bot).
@@coloranalist5193 Looking back a year after this comment was made, I believe the data used as a source for some of these models has been brought up to date. Honestly I imagine if the exercise were to be repeated the results would be surprisingly accurate. I got the opportunity over the last year to actually work with LLMs and AI, and I am extremely impressed.
You explain this code in non programming terms, and explain it as if you are yoda for x in range(0, 3): print("We're on time %d" %(x)) ChatGPT Mmm, code we have. Understand it, you wish. Hmmm. In simple terms, this code is like a wise Jedi going through a series of tasks. The "for x in range(0, 3)" part is like saying, "Let's do something three times." Then, "print("We're on time %d" %(x))" is like the Jedi reporting back each time they finish a task. So, it would go something like this, "We're on time 0... We're on time 1... We're on time 2..." Yes, like that, it is. Understand, do you? Hmmm. 😂
I abuse the living sht out of chatGPT to code better, really helped out a lot when reverse engineering different scripts and translating to different languages!
haha, its funny that you criticize ChatGPT for not being able to write/fix code, as every attempt i made, asking it to optimize or fix code, has broken it..
Thanks for the Great video Marcus and Nice to see this positive outlook on ChatGPT. Separately For people who want to be practical straight away I’d recommend having a look at the book ‘Learn To Code By Solving Problems’ by Daniel Zingaro. It introduces one concept at a time then gets you problem solving. Which is different to the normal style of programming book.
