- Inclined.ai
- Posts
- 🤖 AI To The Danger Zone
🤖 AI To The Danger Zone
PLUS: News Outlets Want To Build A Bot You Could Trust
What's up? You're reading Inclined AI. The newsletter equivalent of vanilla ice cream in a sundae—the cornerstone of all the fun stuff going on
Here's the headline cherries on top:
AI becomes the wingman for the Air Force
News outlets try a new approach to AI
Alibaba launches an AI model
Midjourney allows editing in an update
Modern AI is taking to the skies in a unique, complex way.
The United States Air Force is experimenting with a fighter jet, XQ-58A Valkyrie, that is pilotless. I know what you’re thinking: So it’s a drone?
a robot fighter jet pilot in a scene from Top Gun —ar 2:1 —chaos 25
Sort of, yes. The Valkyrie jets are the next generation of drones that make decisions in a new way. The AI is even capable of flying as a wingman for human pilots.
But flying with an AI comes with a lot of baggage.
Some of that is good, and some of it is…tricky.
On the one hand, Valkyrie shows us that the US military can expand and update its weapons infrastructure in a way that puts fewer human lives on the line.
On the other hand, these jets can take action on their own if we let them. Right now, a board must approve strikes pre-mission, but the Air Force says that could be a temporary measure.
There’s no human on trigger control, unlike drones. That fact can leave you nervous, but there’s a reason they’re pushing forward with this concept.
Right now, the Air Force loves this program because it’s “affordable mass.”
They don’t see these as simple stealth drones. It’s an affordable way to upgrade.
The Valkyrie program costs them $25m for the most expensive model, which is cheap compared to the $80m it takes to make the most affordable fighter jet with a human pilot.
The USAF can afford to make tons of these and toss them at an enemy.
It changes the way we think about warfare and how we handle risk.
The hope is that the shift to “mass” over soldiers highlights the value of human life and the need to avoid deadly conflicts.
Generative AI has landed in the data domain with a bang.
Einblick is the AI-native data notebook that can write and fix code, create beautiful charts, build models, and much, much more–all with as little as one sentence. (pretty cool right?) This is all made possible by their AI agent, Einblick Prompt.
We all know the data analytics and data science process can be a real slog. From debugging syntax errors, to searching through documentation and Stack Overflow threads…it can be a real pain. That’s all before you even get to extracting meaningful insights from your data!
Luckily, Einblick Prompt removes all the tedious parts of the process by leveraging generative AI. Since Prompt is built into every data notebook, you don’t have to copy-paste any code, or switch between tabs. The code is generated straight into the notebook, and you can run it and test it immediately.
And the best part?
Prompt is context-aware. That’s right. You don’t have to type paragraphs of information about your data to get decent results. Prompt will infer which table to use, which libraries to import, and what data types are in your dataset. Plus, if there are any errors, you can debug instantly with the “Fix with Prompt” button.
Don’t take our word for it though. Try it out yourself–it’s free. (You don’t even need your own OpenAI API key–Einblick’s got you covered.)
Media outlets can’t figure out how to integrate AI into their sites.
Blog bots may suck at their job, but companies' profit margin on AI-written articles is enough to justify the decision.
But what’s next for AI in the media?
How about AI chatbots trained on an outlet’s articles? It sounds obvious, but there’s one massive wall in the way of making this work, and its name is Google.
Imagine that you’re on Macworld’s website. How did you get there, and why? Odds are you googled something about Apple, and their article showed up.
That’s why you can laugh at this new-fangled idea of AI chat search built into outlets.
No one clicks a chat icon on news sites to ask simple questions. It’s a learned behavior that you must prompt users to do, but who trusts that?
In the meantime, these companies will spend a ton of money trying to make these bots accurate and reliable for a feature no one wants.
Outlets fumbled the blog bot movement, and now they’re moving the wrong way.
I know we’re all still exploring AI, and anything is possible with this technology, but that doesn’t change facts. No one will go to PCWorld with search queries, and if they did, the tool doesn’t work well enough to justify the release.
What a poor decision.
🎞️ VFX artists in Hollywood show the benefits of adopting AI without disruption
🤖 The trouble with web scraping is coming to the surface, and news outlets are tired of it
🐾 Alibaba launches a new AI model that understands images
🎓 ChatGPT can help college applicants better their chances
🗣️ Poland’s PM was DeepFaked in a new election ad that did not go over well
📄 A new study highlights the potential for LLMs to leak private information
⛵️ Midjourney updates let you request edits to images you generate
🪖 AI is growing its involvement in warfare as more countries look for ways to include it
Check Out Our Sunday Edition
You can dive into more AI news and topics with us every week by subscribing to our premium edition.
We’ve written about the following:
If you’re not already subscribed, that’s okay. We’re offering a free 7-day trial so that you can read this one. That’s how excited I am to post it.
MindPal - discover more, search less (link)
Lily - personal AI therapist (link)
Pipio - personalized video content w/ realistic AI avatars (link)
Almo - new age customer support (link)
Levi - stunning AI-powered websites in seconds (link)
Meme Creator - craft memes w/ an AI (link)
Layla - private AI that runs completely offline (link)
Triplay - playful AI journey planner (link)
Real Lego Cities
This is apparently Toronto. Do any locals want to tell me how accurate this is?
- That’s it for today. I hope you enjoyed the latest edition of inclined.ai - Davis.