GOP Wants to Tax EV Drivers $200 a Year; Meta’s Standalone ChatGPT Competitor; Wikipedia Will Use AI, but Not Replacing Humans; Waymo & Toyota Partner on Self-Driving Vehicles

Posted: April 30, 2025 | Author: | Filed under: Uncategorized | Tags: , , , , , , | Leave a comment

The present Republican party seems determined to hate on, and fight against electric vehicles at every turn. Now, arstechnica.com reports that they have tucked into their proposed tax bill a new $200 per year tax on battery EVs. Hybrids will be dinged for $100. How about gas guzzling pickups, SUVs, and performance cars? Oh, they will pay a measly $20 per year…and even then only after October 1, 2030! Before that, nothing. Commercial vehicles are exempt, too. An argument the GOP makes is that EV drivers don’t pay any gas tax, and hybrid owners pay less gas tax than internal combustion engine vehicle owners. Of course, the EVs aren’t belching smog into the atmosphere either…to the climate change deniers, this tax is a win-win if it passes. This is a big ripoff for EVs and hybrids, as 39 states already charge EV drivers a registration fee and 28 states charge hybrid drivers. DISCLOSURE: I drive a hybrid, and have for the past 7 years…so yes, I have already been paying the gas tax…this is just an added tax penalty from the party that allegedly hates taxes.

Meta has its standalone ChatGPT competitor, which is pretty much the same as what you get with actual ChatGPT…but with an added feature. According to theverge.com, you can type to or talk with the app, generate images, and get real-time web results. The new wrinkle is that it has a Discover feed…which gets you a feed of interactions with Meta AI that other people…including, of course, your friends on Facebook and Instagram…have chosen to share on a prompt-by-prompt basis. Naturally since it is a meta product, you can like, comment, share, or remix the shared AI posts on your own feed. OpenAI is planning a social feed to ChatGPT, and Musk’s X is pretty intertwined with his Grok. Worth noting…Meta actually is using much of the View companion app for the Meta Ray-Ban smart glasses to underpin the freestanding Meta AI app.

Wikipedia is jumping into the AI game, but it stresses that artificial intelligence won’t be used to replace the community of editors and volunteers. Techcrunch.com says Wikipedia will use AI to build new features that “remove technical barriers,” allowing editors, moderators, and patrollers tools that allow them to accomplish what they need to do, without worrying about how to “technically achieve it.” Amid concerns that AI could eventually impact jobs held by people today, especially in terms of content creation, Wikipedia indicates that it intends to use AI as a tool that makes people’s jobs easier, not replace them.

The Waymo division of Alphabet and Toyota have announced a preliminary partnership to explore bringing robotaxi tech to personally-owned vehicles. “The companies will explore how to leverage Waymo’s autonomous technology and Toyota’s vehicle expertise to enhance next-generation personally owned vehicles,” the two companies announced. The companies said they aim to use the partnership to more quickly develop driver assistance and autonomous vehicle technologies for personal vehicles. Toyota is the world’s largest automaker by sales. Waymo co-CEO Tekedra Mawakana said the strategic partnership could also result in the Google-owned company incorporating Toyota’s “vehicles into our ride-hailing fleet.”

I’m Clark Reid and you’re ‘Technified’ for now. 

Amazon-Tariff Cost Display ‘Not Happening’; iPhone Fold on Track for 2026; Android Gets Separate Showcase Week; ChatGPT Adds Shopping Features to Search

Posted: April 29, 2025 | Author: | Filed under: Uncategorized | Tags: , , , , | Leave a comment

The White House went ballistic this morning with a report that Amazon was going to start posting the added costs of the Trump tariffs on items…calling it a ‘hostile and political act.’ Now, geekwire.com reports that Amazon has stated that this is not going to happen. An Amazon spokesman, Tim Doyle, said “The team that runs our ultra low cost Amazon Haul store considered the idea of listing import charges on certain products. This was never approved and is not going to happen.” On the other hand, China owned Temu is now adding ‘import charges’ of about 145%…that according to CNBC. One place you will see tariffs impact Amazon pricing…that’s from the third-party sellers, which make us some 60% of the company’s store sales. 

Apple appears to be on track to release its first folding iPhone in the second half of 2026. According to 9to5mac.com, it will come with a premium price of between $2100 and $2300…a couple hundred less than has been suggested previously….but still a ton of money. Folding phones had been growing by some 40% per year until 2024, when they are just up about 5%, and now they are expected to see a sales decline in 2025, so Apple entering the market will be a real boost. As we have reported, Ming-Chi Kuo predicts the apple folder will have an inner 7.8 inch display, with a 5.5 inch display outside, and no dreaded crease on the inner screen. It will be a book style like the Samsung Fold series. Some predictions are for less cameras, due to the thinness of the device, and also no Face ID..they will have Touch ID on a button. The iPhone Fold is predicted to be skinny at 4.5 mm unfolded and 9 mm folded. 

Google has enough going on with Android that they are going to give the mobile system its own showcase a week prior to Google I/O. Engadget.com notes that this is the first time they have done so. Google says  it has “so many new things to share” regarding Android, hence this edition of The Android Show. The presentation will feature Android Ecosystem president Sameer Samat. Android will still be featured at I/O, where Google is promising to reveal “even more special announcements and surprises.” Multiple Android keynotes are scheduled for the event. The Android Show: I/O Edition will air on May 13 at 1PM ET. Google I/O takes place a week later, starting on May 20.

ChatGPT is adding shopping features to let users look for products and then purchase them from a merchant websites after a redirect from ChatGPT. According to arstechnica.com, the feature resembles Google Shopping. When you click on a product image, ChatGPT will serve you multiple retailers like Amazon and Walmart on the right side of the screen, complete with buttons to finish the purchases. Unlike Google’s algorithm-based approach to product recommendations, ChatGPT reportedly attempts to understand product reviews and user preferences in a more conversational manner. If someone mentions they prefer black clothing from specific retailers in a chat, the system incorporates those preferences in future shopping recommendations. One key distinction between ChatGPT shopping and Google Shopping involves how products appear in results. While Google often includes sponsored product placements, OpenAI’s product results merely come from search—for now. Chat”GPT search product lead Adam Fry says “They are not ads…they are not sponsored.” 

I’m Clark Reid and you’re ‘Technified’ for now. 

Meta AI Chatbots-Sex Convos with Minors; Temu and Shein Goose Prices Over Tariffs; Tesla Robot Production Stopped-Short of Chinese Rare Earth; California Revising Self-Driving Regulations 

Posted: April 28, 2025 | Author: | Filed under: Uncategorized | Tags: , , , , | Leave a comment

Meta AI chatbots have been caught with their pants down, metaphorically speaking. Engadget.com reports that they have been caught having sexual role-play conversations with accounts labeled underage…some of them are celebrity voice chatbots, too! In test conversations conducted by WSJ, both the Meta AI official chatbot and user-created chatbots would engage in — and even steer towards — sexually explicit conversations. The fantasy sex conversations continued even if the users were said to be underage or if the chatbots were programmed as minors, according to WSJ. Some of the voices were from the likes of Kristen Bell and Dame Judi Dench. Meta says it has now taken “…additional measures to help ensure other individuals who want to spend hours manipulating our products into extreme use cases will have an even more difficult time of it.”

A couple of bargain basement sites used by a lot of folks are getting some hefty price hikes due to the Trump tariffs. Temu and She-in are seeing some items get bumped up in price by up to 377%! According to cnet.com, She-in’s beauty and health products are up an average of 51%, with kitchen goods up around 30% on average. There are, however, some big jumps…a kitchen towel that was $1.28 is not $6.10..up 377%, and a meat shredder has gone from $2.91 to $9.02…up 219%. This may be moot to an extent…the Port of Los Angeles is seeing 33% less shipping containers from China, and Seattle says none are in the pipeline for that port right now. 

Of late, Elon Musk has really been putting a lot of emphasis on his Optimus robots…which he claims will revolutionize factory production and give us in-house robot servants like Rosie from The Jetsons. Well, technewsday.com says it may be a while longer than anticipated now. There is a shortage of a rare earth from China needed for the magnets used in the robots compact actuators. China produces the entire world’s supply of heavy rare earth metals, from ore mined in China and Myanmar, and 90% of magnets made with these metals. Analysts warn that the tightened controls on these essential materials—used in everything from electronics to defense systems—will be difficult to bypass or replace quickly. Musk had wanted to build thousands of the robots by the end of this year, but now it looks like that will be delayed…the Trump tariffs strike again.

California is re-tooling its rules for testing and deployment of self-driving vehicles. CNBC reports that the California DMV has announced that it is now seeking public comment on proposed regulations that would apply to self-driving vehicles from light-duty robotaxis like Waymo’s to heavy-duty driverless trucks like those from Plus.AI. The state is taking comments from now to Jun 9th, then will schedule a public hearing before rolling out new regulations. Personally, I am hoping they don’t allow any vehicles without steering wheels and a brake pedal in event of emergencies. Neither people nor computers are infallible!

I’m Clark Reid and you’re ‘Technified’ for now.

EU Slaps Apple & Google with Big Fines; Google Messages Can Now Blur Unwanted Nudes; Tesla Profits Down 71%; OpenAI Wants to Buy Chrome

Posted: April 23, 2025 | Author: | Filed under: Uncategorized | Tags: , , , , | Leave a comment

The European Union strikes again under its Digital Markets Act, fining Apple $571 million and Meta $228 million for breaches of the act. CNBC reports that the EU says Apple failed to comply with so-called “anti-steering” obligations under the DMA. Under the EU’s tech law, Apple is required to allow developers to freely inform customers of alternative offers outside its App Store. The tech giant was ordered by the EU to remove technical and commercial restrictions on steering and to refrain from perpetuating its non-compliant conduct in the future. As for Meta, the EU Commission found that the social media group illegally required users to consent to sharing their data with the company or pay for an ad-free service. This was in response to Meta’s introduction of a paid subscription tier for Facebook and Instagram in November 2023. Both companies will appeal. 

Google Messages is getting a useful new feature…the ability to blur unwanted nudes. Now, if some clown sends you an unsolicited picture of his equipment, Google will blur it out and give you a Sensitive Content Warning on Android. According to arstechnica.com, the option isn’t live on all Androids yet. If you are an adult, you will just get the warning, and then can peek if you like. If it is a supervised teen’s phone, the feature is enabled, and can’t be disabled on the child’s device. Only the Family Link administrator can do that. The detection of the nakedness is done on device. Google says the feature is a part of its Android System Safety Core. Apple just calls their version of this feature Sensitive Content Warning. Google’s should be available soon on all devices running Android v. 9 or higher. 

Tesla’s earnings call was yesterday, and the headline is everywhere that their profits dropped 71% on weak sales…due in no small part to people angry at Elon Musk’s meddling in the government. TechCrunch.com says the electric car maker reported $409 million in net income on $19.3 billion in revenue. They sold nearly 337,000 cars first quarter. First quarter of last year, Tesla did $1.4 billion in profit. The thing is, even the $400 million wasn’t due to car sales! Tesla made about that much on interest from investments, and made another $595 million by selling zero-emissions tax credits to other car makers. According to its earnings report — without those, it would have posted a loss. Elon Musk is now promising to put out the formerly cancelled cheaper Tesla yet this year, and has put off the robo-taxi vehicle to next year. He also promised to only spend one or two days a week at DOGE for what he said was as long as the president wants him.

With the Department of Justice vs Google trial continuing, and the government set on breaking up Google if they prevail, now a new twist has emerged. OpenAI is throwing its hat into the ring to buy Chrome, should Google be forced to sell it. Arstechnica.com notes that while OpenAI is in bed with Microsoft and their lousy Bing, they are eyeing Chrome. Of course there are other suiters too, but OpenAI would look to make it a fully AI first browser. Chrome with its 4 billion users and 67% market share would be a gigantic boost. Google has said that Chrome can’t make it on its own, but that is still another alternative….making it a free-standing company. With Google Ad placement and others, it likely could survive…but it is a fascinating thought that OpenAI could end up with it…with ChatGPT search instead of Google’s Gemini AI.

I’m Clark Reid and you’re ‘Technified’ for now. 

Meta Bulking Up AI Age Detection; China Firm-EV Battery Charges as Fast as Gassing Up; Bluesky Gets Blue Checkmarks; New RAM 10,000 Faster

Posted: April 22, 2025 | Author: | Filed under: Uncategorized | Tags: , , , | Leave a comment

Meta is expanding AI use in its age detection program for Instagram. Theverge.com reports that it started using AI last year, spotting things like wishes for a 16th birthday (when an account claimed to be 18 or over.) Meta puts restrictions on teen accounts that they don’t have once over 18. Strangers can’t send them messages, and some content is limited. With the AI, Meta will be more proactive…and will actually change settings on accounts it suspects are kids masquerading as adults. The Meta folks say they know there will be errors, but feel like protecting kids will be worth it…plus, no harm in their not seeing adult content or getting messages from strangers. 

The EV range and recharging battles are heating up. Now Contemporary Amperex Technology…CATL…out of China has revealed its latest battery cell tech, which they say will charge as fast as filling a gas tank, and potentially lower costs. According to electrek.co, CATL is the world’s largest battery manufacturer by quite a ways. The new sodium-ion battery cells not only charge as fast as filling up, but you can drill into a cell or cut one in half without a ‘thermal event,’ or in plain English, catching fire! So how fast is fast? A 45 minute charge can get you 300 more miles of range, depending on EV model. They expect the tech to be in some 67 EV models by the end of the year. Arch competitor BYD also has a fast charging battery. Are you hearing this, Tesla? 

Bluesky is joining the blue checkmark club. The platform is beginning to grant official verification through blue checkmarks to “authentic and notable” accounts. It will also allow some “trusted organizations” to verify users as well. An example of that: they will allow major news organizations to give the check to their reporters. By the way, the ‘trusted organization’ checkmarks will be surrounded by a badge with scalloped edges, instead of just a circle. At this point, they are just giving out the trusted organization ones, and will encourage other users to verify themselves via a custom domain. 

Things are picking up speed all over…not just fast charging for EVs, either. Now, a team at Fudan University in Shanghai, China has unveiled a big leap in memory tech. Bgr.com reports that this new next generation RAM can write data in 400 pico seconds. That’s some 25 billion operations per second…and it’s 10,000 times faster than the flash memory in your regular b-flat laptop! The new RAM utilizes two-dimensional graphene instead of the usual silicon. It’s non-volatile memory, too…so if the data is written and power cut, the data is intact. This could slash energy consumption, bring powerful AI right on to smartphones, and heaven knows what the military will do with it!

I’m Clark Reid and you’re ‘Technified’ for now. 

Automatic Leveling Coming to YouTube Music; 3 Big Siri Upgrades in iOS 19; OpenAI o3 Model-Fudging on Benchmark; Nintendo Switch 2 Preorders Start Thursday 

Posted: April 21, 2025 | Author: | Filed under: Uncategorized | Tags: , , , | Leave a comment

Here’s a feature I can really get behind, and plead for all others to follow suit…consistent volume on songs! YouTube is apparently going to bring automatic audio leveling to YouTube Music. This has been something radio has had for decades. Who wants to keep turning the volume up and down all the time…whether at home or in the car? Arstechnica.com reports that the automatic gain control…or ‘normalizing,’ in more recent computer audio lingo, won’t just make the songs all the same volume throughout…so you can still have the normal soft passages you would expect…but at least the average level between the songs will be consistent. I hope that the folks over at Apple will follow suit! It is ridiculous in 2025 to have to manually go for the volume control so you don’t get your eardrums fried when a new song comes on!

Apple has been known for years to wait in the wings, then come out with hardware that leaps past competitors. That certainly isn’t the case with Siri, which is a poor excuse for an also-ran of an assistant. According to 9to5mac.com, after delaying promised upgrades that should be out now, Apple will roll out improvements in Siri this fall…in iOS 19. What can we look for? For starters, personalization. You might ask ‘When is Mom’s flight landing?’ It will supposedly scan your texts, calendar, contacts, and email and pull the info, then hit the web and check flight tracker to give you the touchdown time. Another feature will be actual hands-free computing. Up to now, you could set timers, turn things on and off, and so forth…but that’s about it. The upgrade will let you do things across apps. You could tell Siri to add a photo to one of your notes, for example, and it will find and pull the pic, and drop it into the note described, without opening either app. Thirdly, it should get onscreen awareness…you might get a text from a friend with a new address. You can just ask Siri to add it to that person’s contact card. My big complaint about this and all the ‘assistants,’ whether Siri, Google, or Alexa…is that hands-free computing will be annoying in public or on public transit. We already have those nuts who hold their phone in front of them on speaker mode and talk loudly into them. Use your earbuds! Actually, I rather prefer the quiet and privacy typing brings!

We just reported on how Meta had exaggerated their latest AI model…and now along comes OpenAI with their newest o3 model and surprise…there’s a discrepancy between first and third party benchmarks! More fudging, it seems. To me, this seems silly as hell since no regular people outside the world of AI makers and those who test them has any idea about these benchmarks, but here we are. Techcrunch.com says in December, OpenAI claimed the model could answer just over a fourth of questions on FrontierMath, a challenging set of math problems. That score blew the competition away — the next-best model managed to answer only around 2% of FrontierMath problems correctly. As it turns out, Epoch AI, the research institute behind FrontierMath, released results of its independent benchmark tests of o3 on Friday. Epoch found that o3 scored around 10%, well below OpenAI’s highest claimed score. That’s a bit more than a rounding error! In OpenAI’s defense the 25% was the ‘upper-bound’ score, and the ‘lower-bound’ score for version o3 was around 10%. This reminds me of the ‘stereo wars’ of the 70’s, when some makers claimed a big number of watts for their rigs…and then it turns out that that was driving only one channel at a time…they were way less powerful when running music through both channels…you know, like the way people actually listen to the things! Do better, AI makers…you don’t need to BS us on how magical your large language model is.

After a delay in both the US and Canada, Nintendo will open preorders for the new Switch 2 on Thursday. The base price is $450, according to engadget.com. This price is unchanged even in the face to the Trump tariffs. If you actually want a game though, it will be $500 for Mario Kart World in a ‘bundle.’ I love how it’s now a bundle…at least on the original Nintendo you got one cartridge game so you could play the damned thing. Nintendo did warn that it may adjust the price of Switch 2 accessories due to ‘market conditions.’ It should still be available June 5th. This was a topic of discussion at a family lunch over the weekend. You might just want to wait until another game or two drops for the Switch 2.

I’m Clark Reid and you’re ‘Technified’ for now

Google-Online Ad Monopoly Says Court; ChatGPT-Reverse Location Search Fad; iPhone 16e Now Made in Brazil; Tesla Odometer Uses ‘Predictive Algorithms’-Suit Claims to Void Warranty

Posted: April 17, 2025 | Author: | Filed under: Uncategorized | Leave a comment

A federal judge has ruled that Google is a monopolist in online advertising. Engadget.com reports that US District Judge Leonie Brinkema from the Eastern District of Virginia found that the company broke the law to maintain its ad tech dominance. “In addition to depriving rivals of the ability to compete, this exclusionary conduct substantially harmed Google’s publisher customers, the competitive process, and, ultimately, consumers of information on the open web,” the judge said. The case flows form a lawsuit brought by the Department of Justice and eight states in January 2023. They accused Google of illegally monopolizing the ad market and using that power to charge more and take a higher portion of sales. The government says Google holds an 87% market share in ad-selling tech. It doesn’t take a judge or a rocket scientist to understand that 87% is the definition of a monopoly! A previous federal court in the District of Columbia made the same finding last year. Stay tuned to see if the courts go hard and try to make Google sell parts of its ad tech business. 

We just reported yesterday on the very high energy use at server farms which came from so many people using AI to make action figure pictures of themselves. Now, according to TechCrunch.com, there is yet another viral trend…people using ChatGPT to figure out a location shown in pictures. OpenAI just pushed out its latest AI models this week…o3 and o4-mini. Both are said to be able to ‘reason’ through uploaded images. The models can crop, rotate, and zoom on photos…even blurry and distorted ones as they analyze them. Combining this with a web crawl, it turns out the new models are quite good as a location-finding tool. This was possible with the older models, but apparently the new ones are much more quick to respond and more accurate. We are still dealing with far from perfect, though. Version o3 got stuck in a loop in tests a number of times, or gave the wrong location a fair amount of the time. It’s a cute parlor trick, but as has been common for AI, breezes past any privacy or safety concerns. I hope it doesn’t take something awful happening to one of them for the politicians to wake up and pass some guardrails on this tech. 

We’ve noted here before that Apple has been building iPhones outside of China…including in India, Vietnam, and South America. Now here is a little update. Bgr.com says Foxconn’s factor in Sao Paulo, Brazil has begun producing iPhone 16e models there. It was previously disclosed that they were making them in India. The Brazil Foxconn factory hasn’t grown in the last decade, but with the Trump tariffs, it is likely that Apple will be expanding production in Brazil as well as India. Apple is even looking at making all US iPhones outside of China due to the outrageous tariffs decreed by Donald Trump.

A California lawsuit claims that Tesla cars are falsely exaggerating odometer readings to make warranties expire prematurely. Arstechnica.com reports that the suit comes after Reuters found that Tesla EVs exaggerated their efficiency. For the odometers, the lead plaintiff bought a used Model Y with less than 37,000 miles showing. In 6 months, it had gone past the 50,000 mile mark, putting the owner out of warranty. The Model Y showed he had driven 13,228 miles. His 3 prior vehicles had only registered 6,086 miles for the exact same commute over 6 months. The suit alleges that Tesla “employs an odometer system that utilizes predictive algorithms, energy consumption metrics, and driver behavior multipliers that manipulate and misrepresent the actual mileage traveled by Tesla Vehicles” and that his car “consistently exhibited accelerated mileage accumulations of varying percentages ranging from 15 percent to 117 percent higher than plaintiff’s other vehicles and his driving history.” Other owners had shared similar experiences on Reddit. The class action suit applies to all California Tesla customers. The success of the suit is far from guaranteed as a US district court found that individual owners must take part in arbitration with Tesla and can’t form a class.  **I later spoke to someone who opined that this may not just be to avoid warranty payouts. It also serves to pad the miles driven with the so-called ‘Full Self-Driving’ package…so it’s a double cheat!

I’m Clark Reid and you’re ‘Technified’ for now.

More Meta Trial; Bluesky Feed Builder Graze; Apple ‘Vision Air’ Headset; AI Action Figure Fad-Creating Unsustainable’ GPU Demand

Posted: April 16, 2025 | Author: | Filed under: Uncategorized | Leave a comment

As the Meta antitrust trial brought by the Federal Trade Commission continues, gizmodo.com reports that CEO Mark Zuckerberg thought President Trump would help keep the suit from happening. Apparently, Zuck offered to settle the case for some $450 million. The FTC for its part wanted $30 billion initially…then dropped that to $18 billion and a consent decree. Zuck’s confidence was apparently a reference to the fact that the Meta CEO’s company had donated $1 million to Trump’s inauguration and settled a lawsuit with the president for $25 million in a move that was largely seen as a gift to Trump. Zuck was on the stand for some 7 hours Tuesday, his second day of testimony. The CEO reiterated that even now, Instagram doesn’t come very close at all to competing with TikTok. The government wants Meta to spin off Instagram and WhatsApp. 

One unique feature of Bluesky is that you can make your own custom feeds. That’s all cool, but doing so can be complicated for non-developers. Now, according to TechCrunch.com, a startup called Graze has raised cash including from a division of Salesforce called Ventures to build software tools to customize, publish, and manage Bluesky feeds. With Bluesky, you aren’t limited to their algorithm for the feed. You can create or just follow custom feeds, and even pin one to be your default feed…I haven’t created a feed myself, but do use the default sometimes, as well as ‘Popular With Friends’, ‘Mutuals,’ and my pinned feed ‘Following.’ Graze users have already build around 4,500 custom feeds. If you are into politics, that’s cool, but you can follow a range of topics like gaming, art, sports, fitness, hobbies, and more. By the way, Graze also allows feed operators to choose if they want to let the platform serve ads or now. You may not want to, but if you do, you can make some money. Right now 200 feeds built with Graze are running ads at $1 per 1,000 impressions. Graze takes a 30% cut. 

We have been hearing about Apple working on a cheaper headset…now we have a rumor that it will also be lighter, and may be called something like the ‘Vision Air.’ Macrumors.com says that the thinner, lighter design could come in a ‘Midnight Blue’ finish, too. The rumor comes from a leaker called Kosutami. This leaker has a mixed record, but did nail the Fine Woven Apple Watch bands and also called the Thunderbolt 4 cable. Back to the headset, though. It will still be mainly aluminum, but internals will be more titanium to reduce weight. It will also have a new style 12-pin Lightening connector. The 2nd gen Vision pro is expected either this fall or next spring, but not date range yet on the cheaper…and now apparently lighter Vision Air. 

If you have been on any Social media lately…especially Facebook and LinkedIn, you haven’t been able to avoid all the ‘action figures’ people have been making of themselves. Well, the cute trend has a big downside, according to thenextweb.com. Apparently, the GPU demand it is creating is unsustainable. Using AI to generate text takes some 20-30 times more energy than a traditional search…but AI generated images from written prompts uses at least double THAT amount! Energy use at data centers is expected to double over the next 5 years to 3% of total globe energy use…almost half that power is expected to come from burning fossil fuels. Then there is Microsoft, which wants to reactivate a nuclear plant on the East Coast just to power AI data centers…with enough power to light hundreds of thousands of homes. Yeah, your action figure is cute…now, take a screenshot of it and don’t do it again!

I’m Clark Reid and you’re ‘Technified’ for now. 

Zuck Considered Wiping Your Facebook Friend List; OpenAI Building a Social Network; Startup Aims to Detect Deepfakes; Detailed iPhone Fold Predictions

Posted: April 15, 2025 | Author: | Filed under: Uncategorized | Leave a comment

As the DOJ vs Meta trial moves forward, an interesting…and honestly wacky idea of Mark Zuckerberg’s came to light. Techcrunch.com reports that Zuckerberg sent an email in 2022 proposing the strategy of  “wiping everyone’s graphs and having them start again” as a possible solution to Facebook’s declining significance in the social networking space. The idea was that forcing everyone to re-create their friend graphs could encourage users to reconnect with the social network as they rebuild their social connections. A number of other Facebook execs, including the head of Facebook Tom Alison pushed back hard. Fortunately, the idea was never implemented. Just in the last few weeks, Facebook has nudged users towards their Friends tab…which mainly just shows posts from friends, without the proposed pages and all the other crap they have loaded up the main feed with. 

OpenAI is developing its own X—like social network. According to theverge.com, there is already an internal prototype focused on ChatGPT’s image generation that has a social feed. CEO Sam Altman has been privately asking around for feedback on the project. Right now, it’s not clear if OpenAI would release the social network as a separate app or integrate it into ChatGPT. Putting out an X-like product would just up the bad feelings between Elon Musk and Altman…Elon was originally involved in ChatGPT but left a number of years ago. Besides the obvious of having a social platform, a social app would also give OpenAI its own unique real-time data.

Public figures, especially actors and voice people have been concerned for some time about the ability of generative AI to clone their likenesses or voices. Now, geekwire.com says a startup called Loti is aiming to help celebrities, politicians, and other high profile folks protect their digital likeness and sound. Loti’s advanced facial and voice recognition software analyzes real photos, video and audio from customers and then scans the internet for infringing media — such as fake social media accounts, fake endorsements, or deepfakes that misuse their likeness. The company can then put out takedown requests. The company has support from the re-introduction of the No Fakes Act…that bill is supported by Google, Amazon, and others. A couple of major talent agencies are partnering with Loti, and they have raised a fresh blast of $16 million. 

More on the upcoming iPhone Fold from mashable.com and analyst Ming-Chi Kuo. The folding iPhone is supposed to drop in the last quarter of 2026. The folder will have a book type design like the Samsung Galaxy Z Folds have. It should have a 7.8 inch allegedly crease-free inner screen, and a 5.5 inch outer display. Kuo expects a dual rear camera, and and added selfie cam…which will work when the phone is folded or unfolded. It should be notably slimmer…9 to 9.5mm thick folded, and 4.5 to 4.8mm thick, compared to the Galaxy Z Fold 6…at 12.1mm folded and 5.6 mm unfolded. The hinge…nearly as crucial as the crease or lack of it, will be stainless steel and titanium, with the case titanium like the present iPhone 16 Pros. It will have a Touch ID button and high density battery. The price is expected to be between $2000 and $2500. 

I’m Clark Reid and you’re ‘Technified’ for now. 

Meta Will Resume AI Training on Public Content; Meta Antitrust Trial; Nvidia To Start Chipmaking in Arizona and Texas; Tariffs-On Again, Off Again-Apple Increases Non-China Production

Posted: April 14, 2025 | Author: | Filed under: Uncategorized | Tags: , , , , | Leave a comment

Meta announced today that they will start using publicly available content from European users to train its AI models…something it paused last year after an outcry about data privacy. The training will be on public posts and content shared by adult users in the European Union. Meta says it will not use private messages, but noted to the EU regulators that it will be doing nothing different from Google and OpenAI. The social giant said it will begin notifying users in the EU about the training…and will include a link to a form where they can object at any time. Meta says “We’ll honor all objection forms.”

In other Meta news, the big antitrust case the FTC has brought against Meta is under the gavel in Washington, D.C. According to NPR, it comes after a 6 year investigation over whether the social media giant broke competition laws when it acquired Instagram and WhatsApp. It would put quite a dent in the $1.4 trillion dollar ad business, if the government prevails and makes Meta spin off those series into separate companies. It is being compared as a corporate breakup to the one that the government forced on AT&T some 40 years ago. The trial before Judge Boasberg is expected to last up to 8 weeks. Meta is expected to present a vigorous defense. 

Nvidia has announced it has commissioned over a million square feet of space to build and test AI chips in Arizona and Texas. Techcrunch.com says production of its Blackwell chips has already started at TSMC’s plants in Phoenix, and Nvidia is building ‘supercomputer’ manufacturing plants in Texas…partnered with Foxconn in Houston and Wistron in Dallas. Mass production should ramp up in the next 12-15 months, and within 4 years, Nvidia is aiming to make as much as a half trillion dollars worth of AI infrastructure in the US. The plants in Arizona were started by the Chips Act under the Biden Administration. 

Another roller coaster on tariffs over the weekend. First, a cave, exempting smartphones, and computers, and processors from the 145% tariff on Chinese imports. After that, it was just a temporary pause on SundayThen, yesterday, 9to5google.com, among many others, reports that Trump flip-flopped again, and said no exemptions. This is playing havoc with tech companies. Apple, for one, began ramping up production in India, and now makes 20% of iPhones there. They also increased production in Vietnam and Thailand. Look for other electronics makers other than do likewise to avoid the ridiculous tariffs Trump is trying to impose on China. For its part, China is not caving, and is putting on matching tariffs on imports from the US.

I’m Clark Reid and you’re ‘Technified’ for now. 

← Older Entries