🕵️ The Zodiac Killer’s cipher is finally solved after 51 years, China uses cheap AI chips to control hypersonic weapons, boosting range

Welcome to AI Logs’ final issue, as I go heads-down into build mode with my startup that’s slowly coming out of stealth mode and yes — of course we’re using and building AI.

Throughout the past six months, I’ve brought you the AI news, provided my perspectives and insights, shared new AI tools and taught how to use them, taught some advanced prompt engineering as well as the basics, and generally have attempted to help bring all our readership up to speed and to the forefront of what’s possible with AI in these rapidly changing times.

The reality is that AI has and will disrupt virtually all areas of our lives and human existence, so knowing what its capabilities and limitations are, knowing what inherent risks and concerns are present and what’s being done to thwart them, and also leveraging the tools to enjoy the benefits they are providing others with are all prudent at this time. In board rooms, behind closed doors, the “powers that be” are discussing all this and making decisions that will affect and impact us all — “we the people,” need to do all we can to be part of those conversations and decisions made.

If I could distill the previous 16 issues to a few nuggets of wisdom, or at least paraphrased key takeaways, in true AI “TLDR” fashion, they’d be:

> Compute is everything, and breakthroughs such as Groq’s LPUs could bring us much cheaper, quicker, and more powerful computing. “Compute is the new gold”.

> AI is the first technology to be its user manual and a dynamic one at that! Just ask it how to use itself!

> The elements of a superior prompt include providing a persona, parameters, formatting, and being conversational. Ask it if it understands and is excited to deliver next-level results. Give it enough context to know what you must do to deliver your expected results. And don’t take no for an answer — when your AI says it can’t do something, chances are it can; you need to figure out how to get it to do so. (These are all oversimplified; review previous issues for more on how to prompt “well.”)

> I believe that everything AI generates is already based on tokens! - needs to be attached to a blockchain with smart contracts to know what’s “real,” what digital chains of custody exist, and to keep AI “on the rails.”

> Large World Models could help us achieve AGI, and we need to pay attention to all emergent models like LNMs, LAMs, LWMs, etc., and LLMs, which are the most pervasive yet.

> The future will be agentic! Watch for agents, autonomous agents, and others who can co-pilot virtually everything you do soon.

I could write many more, but that’s what the archives are for! I’ve enjoyed writing this and sharing my AI thoughts and knowledge with you all, and I have appreciated the positive feedback and support. Now that this issue has been written, I shall fade to black into my lab to build, test, and deploy tech that I think might help change the world and raise humanity's vibration. See you on the other side!

I could write many more, but that’s what the archives are for! I’ve enjoyed writing this and sharing my AI thoughts and knowledge with you all, and have appreciated the positive feedback and support. Now that this issue has been written, I shall fade to black into my lab to build, test, and deploy tech that I think might help change the world and raise the vibration of humanity. See you on the other side!

Did a friend forward this e-mail to you?

IE₊ SUPPORT INTERESTING ENGINEERING
_{Invest In Science And Engineering}

Enjoy exclusive access to the forefront of AI content, highlighting trends and news that shape the future. Join a community passionate about AI, delve into the latest AI breakthroughs, and be informed with our AI-focused weekly premium newsletters. With IE+, AI reporting goes beyond the ordinary - and it is Ad-Free.

NEWS

• 🔫 China uses cheap AI chips to control hypersonic weapons, boosting range

  The chip of choice, an NVIDIA Jetson TX2i, cost only a few hundred bucks and is widely available to buy online.
  

• 📹 ‘World’s 1st’ paintball firing security cam offers teargas pellet upgrade

  The company claims that the world’s first smart paintball security system will set a new standard for active home protection.
  

• 🎮 Video gamers outsmart AI algorithms in mapping gut bacteria

  ‘Borderlands 3’ players cracked a complex puzzle about gut bacteria that even advanced algorithms couldn’t.

MUST READ

The Zodiac Killer’s cipher is finally solved after 51 years

In 1969, a serial killer terrorized Northern California with a very public yet covert persona. The Zodiac Killer murdered five people, though some called his methods “boring,” and rumors spin that number into more alarming figures such as 49.

Notably, he even threatened the media that he would keep killing if they didn’t publish his correspondences.

However, one of the most intriguing things about the Zodiac Killer was his ciphers. As per reports, the killer had mailed ciphers or encrypted messages.

PROMPT OF THE WEEK

This is a cool one for the music generation. Suno is still my go-to, and I am blown away by it; now on its V3, but a newcomer — Udio — is also great, and Stability just released their Stable Audio, which is promising as well. All are text-to-music AIs that generate lyrics or accept your own, output virtually any genre, and make amazing music out of thin air. But most users don’t feel they have full control of the input — that’s where this super prompt comes into play. Using brackets, you can tell it to do things like [drop], [guitar solo], [backing vocals], [tempo change], etc., and this gives you control over the next order of magnitude. I dare say it’s a “game changer” and takes using these tools to the next level!

AI PICTURE OF THE WEEK

TUTORIAL

How to edit images on ChatGPT/DALL•E:

Once you’ve had DALL•E generate an image, click on the image itself to make it full screen; at this point, you will see the “edit” button on the bottom right, underneath the generation. It will allow you to type desired edits on mobile and desktop, but on the browser version, you can select regions to micro-edit and keep editing until you achieve the desired results.

This is a great way to create slides, infographics, and still images to animate using a text-to-video platform like Runway ML, Pika, etc. (both featured in previous AI Logs issues).

TOOLS OF THE WEEK

🗣️ Adobe Premiere: Simply, WOW. They are bringing various text-to-video tools like Pika, Runway, and Sora! There is seemingly nothing Adobe Premier CAN’T do in the world of design and creation, and it is DEFINITELY a tool to play with if you haven’t before or haven’t recently.

🎵 Udio: I love the name as it’s “studio without the st” and “audio without the a”, but it’s more than just a clever name. Previously, the clear and present winner in the AI audio/music generation space was Suno, but Udio brought some healthy competition! This text-to-music AI is free, easy to use, and has fantastic outputs. Anyone who loves music needs to try this tool (and Suno).

🧐 Reka Core: This is a new multimodal LLM that can actually process video inputs (which most others, even ChatGPT and Gemini, can’t yet). It seems to have cracked the code on browsing the web in a way we, the users, want, need, and demand, and it is performing incredibly well in benchmark testing. Reka has quietly been around for a while, but Core is their newest model, showing much promise. I highly recommend checking it out.

💻 Uizard: AI has got pretty good at coding (although you still need to know the basics like setting up local dev environments, being able to read over the code, deploy it, etc. yet) and tools like Open Devin, Devika, and even just ChatGPT and Gemini have got pretty good at writing code in general, but what about the front end? Well, that’s what Uizard has solved for! This intuitive AI-driven UI creator actually generates pretty amazing IU with a few pushes of a button and is highly customizable. It is a potential game changer, too, but it will certainly raise some eyebrows in the UI/UX community.