- NATURAL 20
- Posts
- NASA ChatGPT Prompt, AI Powered MMO and John Romero AI Powered Game Design
NASA ChatGPT Prompt, AI Powered MMO and John Romero AI Powered Game Design
White House J's AI executive order, focusing on bolstering AI safety and Neural MMO Challenge and its potential to reshape AI's role in gaming.
Wes Roth
October 30, 2023
Today:
NASA ChatGPT Prompt, AI Powered MMO and John Romero AI Powered Game Design
NASA's put out this mega-prompt thing for GPT-4. It's like having a chat with someone who helps you refine your ideas—think of it as having a brainstorming buddy in your pocket. You throw an idea their way, and they help you mold it, ask questions, and make it solid. Folks have been finding it pretty handy.
In other news, there's a new challenge called the Neural MMO. Think about it like World of Warcraft, but instead of humans, it's AIs playing the game. They'll need to wear helmets, armors, trade items, and can even bag real cash prizes—like 20 grand! The whole deal is to train a bunch of these AI agents to finish tasks in the game, kinda like training robots in a simulation. The catch? These skills they pick up could be used in the real world too.
G7 to agree AI code of conduct for companies
The G7, a group of seven major industrial nations, are getting together on Monday to lay down some ground rules for companies playing with advanced AI tech. They're doing this because they're worried about what can go wrong if AI's misused. This isn't some mandatory law, but more like "best practices" advice. It’s a way to make sure AI is safe, trustworthy, and being used for the right reasons.
Companies should check for any issues throughout AI's use, fix any problems, and be open about what their AI can and can't do. They should also beef up their security. Europe's been pretty strict about regulating AI, while the US, Japan, and some Asian countries are letting things slide a bit to help their economies grow. The big tech boss in Europe said that this code of conduct is like a safety net until they come up with stricter rules.
A group behind Stable Diffusion wants to open source emotion-detecting AI
Remember when Alexa started picking up on our moods? Well, now there's a push to make "empathetic" AI available to all developers for free. The nonprofit LAION, which provided some data to help develop emotion-sensing in tech, just launched the Open Empathic project. Their mission? Make AI not just smart, but also emotionally intelligent.
LAION is rallying folks to contribute audio clips to help the AI understand our feelings better. Christoph Schuhmann, a bigwig at LAION, sees a future where robots can comfort us, and virtual assistants can tell when we're feeling blue. But not everyone's on board. Some experts warn that trying to make AI recognize emotions is shaky science. Different cultures express emotions differently, and there's a real risk of getting it wrong. And don't even get me started on bias. Past emotion-detecting tech hasn't had the best track record, sometimes misreading based on race, gender, or other factors.
LAION believes in their open community approach, hoping to keep things transparent and above board. But with big rewards come big risks. Whether this "feelings AI" is the future or a flop, we'll just have to wait and see.
OpenAI, Google and a ‘digital anthropologist’: the UN forms a high-level board to explore AI governance
The UN's diving deep into the AI game. They've just announced a new advisory board with 39 bigwigs from government, academia, and industry, aimed at guiding AI's global governance. This group's gonna act as a bridge for all the UN's AI initiatives. They've been chinwagging with the big players in the AI world and are looking to drop some recommendations by the summer of 2024.
This board isn't just about the scary bits of AI, like misinformation threats, but also its potential for good, like boosting public health and education. The UN's big cheese, António Guterres, said AI has huge potential to boost global progress, especially for developing countries. Still, there are folks like Gary Marcus who are wary, emphasizing the challenges of controlling AI. The board's members? A mixed bag from all over, including folks from Google, Microsoft, and even a "digital anthropologist."
Content Credentials That Label AI-Generated Images Are Coming To Mobile Phones And Cameras
Big companies like Qualcomm and Leica are adding new tech to phones and cameras that can tell if a picture is real or made by a computer. With lots of fake images online, this aims to help people figure out what's true.
These devices will add special details to photos about where and when they were taken that can't be easily changed. Qualcomm's latest phone chips and Leica's new camera will tag photos with this info. This is part of a bigger plan by tech giants like Adobe, Microsoft, and others to set rules for proving if media is legit.
While not perfect, it's better than current methods. Qualcomm's tech even notes changes made to an image and where it came from. The goal is to help folks trust real photos online and spot the fakes. Though, not all devices and apps might use or show these details, Qualcomm wants everyone on board in the future.
The Godmother of AI Calls for US ‘Moonshot’ Investment in the Technology
Fei-Fei Li, a big deal in AI, told President Biden the US needs to seriously up its game in funding AI research and public projects. She’s worried that universities and public institutions are missing out on the AI action because they don’t have the big bucks like private companies. Even with open-source models popping up, it's still a tough game for these public players to create top-notch AI. Li is all for talent moving between academia and industry, but she doesn’t want all the good stuff just happening in big tech.
In her chat with President Biden, she pushed for a bold investment in AI, kind of like the US moon landing back in the day. She’s jazzed about the potential of AI in various fields, from medicine to law, and she’s seen some messiness in tech's impact on society, like the Google Project Maven drama.
Li doesn't lose sleep over AI taking over the world anytime soon. She thinks we should focus on handling the immediate challenges AI brings to the table, rather than getting caught up in doomsday scenarios.
Why companies investing in generative AI want OpenAI alternatives
OpenAI's got this fancy chatbot called GPT-4 that's kinda top dog right now. It's huge, smart, and can talk about everything from baseball stats to how to buff up your resume. But here's the thing: other big companies and some up-and-comers are stepping into the ring. We're talking giants like Google and Meta, plus some new kids on the block like Anthropic and Mosaic.
Why's this happening? A couple reasons. First, businesses don't wanna put all their eggs in one basket. Like, Cisco, that big tech company, doesn't just use one model. They mix and match based on what job they're doing. For some stuff, they go with OpenAI, but for other things, they might use a different model or even make their own.
Next, some companies, especially the little guys just starting out, are trying to save some cash. Training these chatbots ain't cheap. So, they're shopping around, looking for deals, and then tweaking these models to fit their needs.
The future of pharmacies: Robots, AI and more
Pharmacies are starting to use more robots and AI to help out with the behind-the-scenes work so pharmacists can focus more on patient care. This shift is happening because pharmacists are burnt out and overworked, and tech can take some of that load off their shoulders.
Pharmacies have been pretty slow to jump on the tech bandwagon, but the COVID-19 pandemic really lit a fire under them to make some changes. Walgreens is building a bunch of "micro-fulfillment centers" across the country where robots fill most of the prescriptions, and Walmart and some other big chains are also getting in on the action.
On the software side, new AI programs are helping pharmacists keep track of what meds patients are taking, manage inventory, and make sure there aren't any bad drug interactions.
Pharmacies are also working on making their websites and phone systems better so people can get their meds without a hassle.
4 Ways AI Is Changing Marketing
Here's the skinny on how AI's shaking up the marketing game:
Getting to Know You: AI digs deep into consumer info. Imagine a salesperson who knows EVERYTHING about a product and every question ever asked about it. It's like having a limitless army of top-notch sales reps. This means better customer experiences and, yeah, some jobs might get replaced.
Personal Touch: With AI, ads get personal. Say you've got a Labradoodle. Companies like PetSmart can shoot you a text, chat about your dog by name, and suggest stuff other Labradoodles in town are loving.
Crystal Ball Vibes: AI's turning into a future-seer. By peeking at social media, it gets what consumers want now and guesses what they'll want next. Marketers can then prep for future demand and jazz up their products.
24/7 Chat Buddies: AI chatbots are becoming the ultimate customer service reps. They're available all day, every day, and they're getting so good, it's hard to tell them apart from real folks.
11 Ways To Use AI In Content Marketing Without Sacrificing Quality
Content marketers are working their tails off! Most B2B folks spend 33 hours a week on content, and that's just the writing part. Add brainstorming, keyword hunting, editing, and everything post-publish, and we're swamped! But here's a ray of sunshine: AI isn't coming for your job. It's just here to give you a hand, be your sidekick, if you use it right.
How?
Topic Research: Use AI to dig up juicy stats, but double-check for accuracy and relevancy.
Case Studies: Need real-world examples? AI can scout some for you.
Ideation: AI can toss around eBook or blog topic ideas. Think of it like the rookie in a brainstorm sesh.
Content Writing & Editing: Use AI to get started with content outlines or even a rough first draft. It's like a jumpstart when you're staring at a blank screen. But remember, it's on you to sprinkle your human touch.
Landing Pages: Can't nail that perfect headline? Let AI draft a few, then pick and tweak!
Editing: AI can proofread your stuff. Spelling, grammar, tone, and even SEO. A digital buddy to bounce ideas off!
Promotion time? AI can whip up snazzy social media posts for you based on your content. Just give it the right direction.
What'd you think of today's edition? |
What are MOST interested in learning about AI?What stories or resources will be most interesting for you to hear about? |
Reply