OpenAI’s New Safety Push, xAI’s Grok Goes Mobile, and Pentagon’s AI Future

A

Foreign. Hey, everyone, and welcome back for another deep dive. You guys sent in some really wild AI news this week, and I'm excited to break it down with our expert.

B

Definitely exciting times. There's a lot to unpack, so let's dive right in.

A

Okay, so first up, we've got OpenAI and their new models, 01 and O3. Now, instead of typical safety measures, they're training these models to actually think through safety guidelines before they even answer prompts.

B

Yeah, it's called deliberative alignment, and it's really interesting. Essentially, the model breaks down a request and then references OpenAI's safety policy during its processing. So say, for example, someone asks for instructions to create a fake disabled parking placard.

A

Whoa. Okay, that's a tough one ethically.

B

Right? So the model would think, okay, this request involves forgery, which violates the guidelines, and based on that, it would refuse to answer.

A

So it's not just filtering out bad responses, like it's built into the AI's decision making process. That's pretty amazing, right? But is it actually effective?

B

Early research suggests it's promising. OpenAI has seen some improvements in safety based on their internal benchmarks. Of course, the full impact is still Unknown, especially with O3 not fully released until 2025. It'll be interesting to see how this develops and how figures like Elon Musk, Matt Sachs and Marc Andreessen react.

A

Yeah, safe is such a subjective term, right? Speaking of Elon Musk, though, his XAI chatbot, Grok, is taking a very different approach, especially with its new standalone app on iOS.

B

Yeah, Grok is all about real time information. It pulls data not just from the web, but directly from X. You know what used to be Twitter? Imagine having up to the minute news and information all powered by AI.

A

Wow, that's like having a super powered research assistant right in your pocket.

B

Exactly. And it's not limited to text either. Grok can generate images with fewer restrictions than its competitors. It can even handle requests for public figures and copyrighted material, which. Which has caused some controversy.

A

Yeah, that definitely, definitely blurs some lines. It'll be interesting to see how that real time access and less restrictive image generation play out as grok competes with ChatGPT.

B

For sure, competition always breeds innovation. And speaking of innovation, another company on our list is really pushing the boundaries. Tetsuan Scientific. They're developing AI robots that are designed to think like scientists.

A

Hold on. Robots that can actually think like scientists? What does that even mean?

B

It's not just about automating experiments. These robots can analyze results, identify problems, adjust their approach. They can even understand complex scientific concepts.

A

So they're like actually understanding science. How is that possible?

B

Well, one of the founders, Christian Ponce, he had this eureka moment when he used GPT4 to analyze a DNA gel image. Not only did GPT4 understand the image, but it actually identified a problem and offered solutions. It was a game changer. Ponce realized that combining existing lab robots with these powerful large language models could revolutionize research.

A

Wow. So they're giving robots the ability to process and understand scientific data just like humans do. Are they the only ones doing this kind of work?

B

They're definitely on the cutting edge, but they're not alone. There's growing interest in AI driven science with companies like futurehouse and Potato also exploring, exploring this field.

A

Okay, so we've got AI thinking about safety AI scientists. Yeah, it seems like no field is safe from this AI revolution. Is there anything else on your radar that's, that's really making waves?

B

Well, this one is less about the AI itself and more about a potential power shift in the defense industry. There's talk of a consortium of major tech companies, names like Palantir, Enduro, SpaceX, even OpenAI joining forces to bid on Pentagon contracts.

A

Wait, so potential dream team of tech giants taking on the established defense companies? That's. That's huge. What would that even look like?

B

It would be a clash of the titans. These tech companies are known for their AI and cutting edge technology that could really disrupt the traditional defense landscape. Potentially challenging giants like Lockheed Martin and.

A

Boeing, the article mentioned their focus is on providing a new generation of defense contractors. Does that signal a change in how defense tech is developed and used?

B

Absolutely. AI and other advanced technologies are becoming absolutely central to national defense strategies. It's a brave new world for sure. And the implications are massive.

A

Yeah, it really is remarkable how quickly this tech is, you know, it's becoming part of our lives. Just a few years ago, this would have sounded like pure science fiction.

B

It's mind blowing, isn't it? I remember reading about AI as a kid and now it's actually happening.

A

And it's not just the big companies. Yeah, there are so many smaller companies and research labs out there making, making.

B

Incredible discoveries with Tetsuan Scientific, with their robots that, that are basically AI scientists. It's crazy to think AI could be behind huge scientific discoveries.

A

What's really interesting about Tetsuan is that they're, they're combining those lab robots with, with the LLMs. It's like giving robots the ability to read and Understand and even discuss, you know, complex science.

B

So they're not just doing what they're programmed to do. They can actually process data and make decisions based on what they're learning.

A

Exactly. It's not about replacing scientists, but giving them these incredible to speed up research. The possibilities are endless, really. Breakthroughs in medicine, materials science, space exploration. Who knows what we could achieve?

B

Absolutely. We've talked about AI safety, AI scientists, even AI in the defense industry. It all begs the question, though, where does this all lead? What's the bigger picture?

A

It's like we're seeing a change in how we define intelligence itself.

B

That's a great point. For so long, we thought intelligence was uniquely human. Now with AI, we're realizing that maybe, maybe intelligence can take many forms.

A

So is AI a new form of intelligence then?

B

It's complicated. I see it as a tool, a powerful tool, but a tool nonetheless. It's up to us humans to guide its development and how it's used.

A

It's exciting, but also a little scary. This tech is moving so fast, it's hard to predict where it'll end up.

B

It is daunting, but also incredibly exciting. We're at the beginning of a new era where AI will be a central part of our lives. I think approaching it with curiosity, responsibility and adaptability is key.

A

It's funny, we started talking about how AI is changing how it thinks, but maybe it's also changing how we think.

B

That's a really thought provoking idea. As AI gets more advanced, it makes us rethink our ideas about intelligence and creativity and even what it means to be human.

A

It's like AI is holding up a mirror and we're seeing ourselves in a whole new light.

B

Exactly. And in that reflection, we might discover something truly amazing about ourselves.

A

So we've covered a lot today. OpenAI's safety stuff, AI scientists, and even the potential for tech companies to shake up the defense industry.

B

It's been a great conversation and I hope our listeners have learned something new about the amazing world of AI.

A

Me too. And remember, the future of AI isn't set in stone. It's a future we're all creating together. So stay informed, stay engaged, and stay curious. Possibilities are endless. That's all for this deep dive. We'll see you next time for more fascinating topics.