Відео

Something that I've literally not seen anyone talk about, is how ALL these LLM's were brute forced into existence. Deepseek is the finessing of a model, and OpenAI could have always done this. They could've distilled the model themselves and released downloadable versions for the public to use. They were greedy and now the market is paying for it.

Nothing like starting out the show with a bit of irony.

Skylink anyone.

So, what measures are being taken to insure the ability to shut these AI down when they go rogue? Just because you're stupid enough to scoff at those that warn you to have caution, does not mean that caution is NOT warranted.

Consultant and specialist would probably be the first profession to suffer. It will stop nepotism hopefully. It's going to be interesting if AI can actually create freedom for people. True open code is freedom of information. Not a mysterious self serving monster with politics, religion or geographic bias as a golden rule.

Nice

If you don't mind me sharing here a short video that covers the DeepSeek, what should be called, Anti-Privacy policy: ua-cam.com/video/s2eVyiNRr_4/v-deo.htmlsi=mAb9C2rROdGGXkz7

Why would China care if their possession of these chips were illegal? Who would police them?

Another reason Nvidia went down was because Trump said he will put a 25%-100% tariff on TSMC chips that Nvidia relies upon

Great. I thought we were trying to not spend money....

2027 Leopold is a smart young man, they never should have fired him

How many people actually watch the movie Terminator? It’s like they are trying to invent something that will end the world.

Does that little dialogue really reflect the way it works?

What input at what cost and how efficient and effective are all about AI future. As consumers, I definately jump for joy for the most efficient and cheapest or even free AI facility. Great work to China.

Humanity's exam is on... and it's a tough one. We're facing challenges on multiple fronts, from climate change and resource depletion to social inequality and global pandemics. It's a complex test that requires us to use our intelligence, creativity, and compassion to find solutions. The outcome will determine the kind of future we create for ourselves and the generations to come.

I have consumed a good bit of ai content and your format, presentation and thoughtful discussion are among the best. Thank you!

OpenAI's economic blueprint: "either we achieve AGI in a year, or the gig is up and Elon better fire up the rocket to Mars." Since AGI will never be born from LLMs+transformers... OpenAI is in quite the pickle, now that its an open-source game and there is no magic bullet to put the salesman genie back in the bottle.

Excited for the AI literacy project! I followed up and watched Open AI's video demo on Operator too. I have to admit I felt overwhelmed with info after all the things I watched last night! And that I was behind in my knowledge, but I reminded myself how much I've learned by watching you guys for so long compared to my peers! And Stargate - a sci-fi name at last!! Long overdue lol

Thank you for the extra episode for this week. I tried to follow the DeepSeek stuff but got lost with it. So I was glad to hear that I was on the right lines with my thinking based on what Paul shared. I did also wonder about the potential for future disruptions so it was good to hear about that too. It's been a crazy amount of AI stuff to wrap our heads around. I am grateful that this podcast exists to help me make sense of it all.

Thanks for making the point that there is no job more important to replace with AI than an AI engineer. It’s arguably the only job they are working day and night to replace, because whoever does it first, wins the race. I don’t understand why more people are not talking about this. For an AI lab, releasing consumer tools and apps is mostly a distraction. They generate some revenue and data, but there is no way that their primary focus is not on building AI SWEs that can build better AI. Not only that, it’s the one job they are best equipped to replace. If they want to build AI doctors, they need to hire human doctors and pay for proprietary medical data and research. However, if they want to build AI SWEs, which they do, they already have the best talent on the payroll.

The infrastructure thing feels wrong. I imagine even Einstein's brain got by on a couple thousand watts. But AGI requires dedicated nuclear power plants? There has to be a better answer. Ever since transformers, AI development has been like a dog that knows one trick: make it bigger. Time to rethink before we dump trillions into this.

Great insights on Zapier and the differentiation between built-in agents within SaaS platforms versus external agentic solutions. I’m curious-when do you plan to cover Salesforce Agentforce? Launched in 2024, it’s highly relevant and arguably a superior approach in this space.

Thank you for the double episode

Big fan of the podcast but disappointed at the Title gents. "2:04 40% of leading Global employers anticipating reducing their Workforce" is NOT the same as "40% of jobs at risk". This is irresponsible as you'll likely be incorrectly quoted and repeated by others.

This video was made just as the release of DeepSeek's R1 rendered OpenAI irrelevant.

No deepseek?

Just in time for DeepSeek to demonstrate that massive data centers aren't needed to train the top LLM and tank AI tech stocks. Is Softbank cursed to make all the worst investments?

Awesome episode as always guys. Maybe meta are up purely for the open source element?

Here's the crazy thing with OAI's Operator. If they would have added projects where you could upload your grocery receipts and it would then know what product to buy if you have something like eggs on your list, that would be impressive. I just don't think they have the proper people steering these developers in the right direction. They're lacking that vision.

Altman, master hype man

guys you need to check mcp claude

Looking forward to the listen

bye bye human jobs

Why does current AI have trouble with making pictures and using correct spelling. I type in words and AI misspells them. Grok, Chat, they all seem to have the problem.

09.09.2025

Going against international law to ”protect US competitiveness”, and throw in China there as an adversary. This is corporatism, also known as fascism.

Alright bois, good game, well played. Maybe we'll have a fair shot in the next game of life.

I don't think anyone appreciates the illuminati photo in the background

What's a Twitter?

OpenAi is showing off Operator right now release imminent

Why think when AI does it for you? Why work when AI does it for you? All you have left are your instincts. You are becoming a monkey again.

bye bye human jobs

they probably wouldn't have suggested some other ai developer if it wasn't some developer other than openai.

How do you square the first quote with the recent announcement of Stargate?

Signed up for the AI Writer session and shared in a few groups last week. Also booked in for the webinar this Friday, thank you for this I'll be interested to hear what you have planned.

Both of you need to stop looking at your monitor, look at the camera... you make viewers feel like stalkers.

I really appreciate a sensible adult discussion.

I think there's going to have to be a push for efficiency with these AI models at some point. It seems that capabilities are not a moat that can't be breached. The models all catch up and surpass each other pretty quickly. So I think the company that dominates the market will be the one that do it for the least cost. This should be good for the economy as a whole, because it will drive efficiency in so many areas that everyone will benefit from.

"Suhree" bugged me the first 10 times I heard you say it that way, but now I listen for it...gonna start pronouncing it that way myself 🤣Your podcast is the best AI cast of the 3-5 I routinely listen to, thank you for putting it out every week!

I really appreciate you both and this podcast. I truly love it. Thank you! Amazing chat as always!