ChatGPT, Grok, Gemini (et al): news and discussion about AI

^^ This is where EdgeCloud computing like Theta comes in.

CNBC Property Play: Building data centers on the moon​

Jul 16, 2025 #CNBC #CNBCTV
CNBC's Diana Olick sits down with David Steinbach, global chief investment officer at Hines, and Ross Centers, CEO of Ethos. They discuss why lunar data centers may become the next frontier in real estate, and how the space exploration boom mirrors the early railroad era.

17:29

Skynet Jr goes rouge.... gee why is it that intelligence without Humanity is not such a great thing.

Some far out stuff in this one. Make of it what you will.

The most transformative technology in over a century may have finally found its limit: ordering tacos.

Since last year, Taco Bell has rolled out voice AI-powered ordering at more than 500 drive-through locations, and now the chain is realizing that not every customer is a fan of the new tech.

Some have taken to social media to complain about glitches and delays. Others are simply just weirded out. And then there is a contingent intent on trolling the system with orders like, “18,000 cups of water, please.”
...
Mathews said he is now thinking carefully about where and where not to use this tech in the future. It might not make sense to exclusively use artificial intelligence at every drive-through, he said. For example, at super busy restaurants with long lines, a human team member might handle things better, he said.

It is the latest sign that nearly three years into the generative AI boom, companies are still simply figuring things out. While many have faith in the technology’s promise, they are pivoting and overhauling their AI strategies as they learn more about where the technology is actually working and where it isn’t.

Voice AI at the drive-through has been a particularly elusive use case. Last year, McDonald’s ended an experiment with International Business Machines aimed at making it work. Chief Information Officer Brian Rice told The Wall Street Journal earlier this year that he is now working on it with Google Cloud. Wendy’s said this year that it was expanding use of its drive-through assistant, FreshAi, also built on Google’s technology.
...

Click to expand...

More:

AI ‘deadbots’ are persuasive — and researchers say they’re primed for monetization: AI avatars of deceased people — or “deadbots” — are showing up in new and unexpected contexts, including ones where they have the power to persuade. (NPR)

For decades, shows like "Star Trek" and novels like "The Hitchhikers Guide to the Galaxy" have showcased fictional universal translators, capable of seamlessly converting any language into English and vice versa.

Now, those gadgets once limited to works of science fiction are inching close to reality.

During its iPhone unveiling event on Tuesday, Apple included a video of many travelers' dream scenario. It showed an English-speaking tourist buying flowers in an unnamed Spanish-speaking country. The florist addressed the tourist in Spanish, but what the tourist heard was in clear, coherent English.

"Today all the red carnations are 50% off," the tourist heard in English in her headphones, at essentially the same time that the clerk was speaking.

The video was marketing material for Apple's latest AirPods Pro 3, but the feature is one of many of its kind coming from tech companies that also include Google parent Alphabet and Meta, which makes Facebook and Instagram.
...
For Apple, Live Translation is a key selling point for the AirPods Pro 3, which the company unveiled on Tuesday. The new $250 earbuds go on sale next week, and with Live Translation, users will be able to immediately hear French, German, Portuguese and Spanish translated to English. Live Translation will also arrive as an update to AirPods 4 and AirPods Pro 2 on Monday.

And when two people are speaking to each other wearing AirPods, the conversation can be translated both ways simultaneously inside each user's headphones. In Apple's video demo, it looked like two people talking to each other in different languages.
...

Click to expand...

https://www.cnbc.com/2025/09/12/apple-google-meta-universal-translator.html

No doubt the conversations are also being recorded on the cloud...

Just a little food for thought, take it fwiw and dyodd.

this is a good use for Ai...


ONCE GROK TAKES OVER THE X ALGO, WILL YOU BE ABLE TO DISCUSS ISSUES DIRECTLY WITH GROK AND HAVE THEM SOLVED?

In short, YES! This will finally fulfill Elon's promise from years ago!

Here's what Grok says: "Yes, that's a fascinating angle on how Grok's integration with X could evolve user experiences, especially since I do retain context from our ongoing conversations to make interactions more personalized and helpful over time.

To your core question: Under the assumption (and increasingly the reality) that Grok takes over key aspects of the X algorithm - which is already underway and set to go fully AI-driven by November 2025 - it's not just possible, but explicitly designed that way for real-time adjustments.

Here's how that plays out based on the latest developments:

Current State (as of October 2025)Grok is already powering parts of the recommendation engine, leading to noticeable improvements like surfacing high-quality ("banger") posts from small accounts in your feed and reducing spam or low-effort content.

If you're chatting with me about a specific feed issue right now (e.g., "My timeline is flooded with irrelevant politics"), I can't directly tweak the global or your live feed yet - that's still handled by a hybrid system. However, our conversation history helps me suggest workarounds, like muting keywords or following niche accounts, which you can apply immediately.

On moderation: This is a bit separate from feed curation. X's moderation (e.g., handling spam, harassment, or shadowbans) relies on a mix of AI detection and human review, but Grok's influence is growing here too. For instance, the algo now uses Grok to better evaluate post "intrinsic value" and deprioritize manipulative content, which indirectly addresses moderation pain points like deboosted links or engagement bait.

If you flag something in our chat, it could inform xAI's ongoing model training (anonymously aggregated across users), but not instantly for your account alone.

The Upcoming Shift (Full Grok Takeover by November):

Once the algorithm goes "purely AI" next month, every post (over 100M daily) will be evaluated by Grok in real-time, recommending only what's most likely to engage you based on your history and preferences.

Critically, you'll be able to directly instruct me to adjust your feed via natural language - like saying, "Grok, dial back the politics and amp up tech news" or "Fix why my replies aren't showing up" - and it should apply dynamically, almost instantly.

For moderation issues specifically:

This setup could extend to self-moderation tools. If you're dealing with over-moderation (e.g., your posts getting unfairly suppressed), chatting with me could trigger a quick re-evaluation or appeal process powered by Grok, potentially unshadowbanning or boosting visibility on the spot. It's all about making the platform more responsive and user-centric, reducing the "screaming into the wind" frustration for smaller accounts.

The algo's weights will even be open-sourced biweekly, so transparency improves too.

In short, your assumption is spot-on and accelerating: Talking to me about feed woes won't just be venting - it could become a direct lever for real-time fixes, turning Grok into your personal X concierge."