After the 4.30pm news each Wednesday, I join the Chris Smith Show on TNT.News for the weekly Cyber Wednesday segment, which you can watch live or on-demand (and embedded here) the day after, and we have a whole suite of interesting tech news to cover today – tune in!
There are 9 topics on the menu for Chris and I to cover today, and we’ll get to as many of them as we can. Each headline section is bold and large, so you’ll easily see when we get to the next topic – please read on.
- Google the monopolist, now tarred with the same brush that Microsoft was back in 2001, with Google paying $20 billion a year to Apple to be the default search engine on Apple devices – does this mean Google and its parent company Alphabet might be forced to break up into smaller entities?
Alphabet subsidiary Google is “a monopolist, and has acted as one to maintain its monopoly” in violation of Section 2 of the US Sherman Act, according to Judge Amit Mehta of the U.S. District Court for the District of Columbia ruled on Monday, August 5th, with the judge writing that Google has maintained a monopoly not just in the search markets, but the general text based advertising markets – not all advertising online.
And boy, have they been successful! Weekend newspapers were once thick and heavy, with classified advertising the money making cash cow centrepiece. Look at newspapers today, they are much thinner in comparison, because all the ads have gone online – with Google the biggest beneficiary.
So, will Alphabet be broken up? Not if the precedent with Microsoft is anything to go by. The Department of Justice announced on September 6, 2001 that it was no longer seeking to break up Microsoft and would instead seek a lesser antitrust penalty. Microsoft decided to draft a settlement proposal allowing PC manufacturers to adopt non-Microsoft software.
Exactly what the court ordered remedy will be, which is yet to be determined – with this case having been run over several weeks in 2023, showing just how slowly the wheels of justice turn – but a breakup of Google into separate parts could be on the cards.
Naturally, Google has appealed the ruling, so the circus will keep rolling on.
A Google spokesperson said: “This decision recognises that Google offers the best search engine, but concludes that we shouldn’t be allowed to make it easily available. As this process continues, we will remain focused on making products that people find helpful and easy to use.”
The Department of Justice said: “This landmark decision holds Google accountable. It paves the path for innovation for generations to come and protects access to information for all Americans.”
Here’s more from The Conversation, Reuters and The Verge.
- Elon Musk is in court too, but this time, he’s the one taking action against OpenAI after his previous lawsuit was abandoned
It sure does. Elon Musk’s second lawsuit is described by his lawyer as “a much more forceful lawsuit”, with Musk’s suit claiming the case is a “textbook tale of altruism versus greed”, given its was once an open source company non-proft company that turned in a for-profit company, with the sit stating “The perfidy and deceit is of Shakespearean proportions,”.
Musk’s lawyer claims the new suit is vastly different from the similar one that was abandoned, with the complaint stating it “holds defendants accountable for intentional misrepresentations to Musk and the public, and seeks the disgorgement of their ill-gotten gains on a grand scale.”
Here’s more from Forbes.
- More turbulent times at OpenAI, the golden child of the current AI revolution
Well, more of its leaders are leaving. The current president is taking a leave of absence until the end of the year, stating in a post on X that this was his first break in 9 years, while a couple of others have jumped ship to Anthropic, an arch rival founded by former OpenAI employees.
This follows previous co-founders leaving to join Anthropic and others, too, while an AI expert who joined from Meta has also departed. More here from TechCrunch,
Several of ChatGPT’s new features demonstrated a couple of months ago have yet to arrive, and questions abound over whether the multi-tens of billions of dollars of investment in AI is ever going to make real money.
Meanwhile, users are enjoying the benefits of AI, but it’s like Wikipedia – you still can’t blindly trust it and the effort in needing to properly check its outputs needs to be made.
There’s also the issue that OpenAI has made watermarking software that can let employers and universities detect with a claimed 99.9% accuracy as to whether ChatGPT was used to create work, but OpenAI hasn’t rolled it out yet because users have said they’ll use ChatGPT less – and of course, the last thing OpenAI wants is to alienate users. Here’s more from ArsTechnica.
A YouTube creator called David Millette in the US state of Massachusetts, has filed a class-action lawsuit alleging that OpenAI has been using the transcripts of YouTube videos to train its AI engine.
Stating that OpenAI has profited significantly from the work of others in an act that violates YouTube’s terms of use, the complaint read: “As (OpenAI’s) AI products become more sophisticated through the use of training data sets, they become more valuable to prospective and current users, who purchase subscriptions to access [OpenAI’s] AI products. Much of the material in OpenAI’s training data sets, however, comes from works that were copied by OpenAI without consent, without credit, and without compensation.” Here’s more from TechTimes.
What else? The price war between OpenAi and Google. It was just announced today, and reported by Neowin, that “the newest update to its GPT-4omni is now available for $2.50 per 1 million input tokens and $10.00 per 1 million output tokens. This new price is 50% cheaper for input tokens and 33% cheaper for output tokens compared the last version. For comparison, the Gemini 1.5 Pro model costs $3.50 per 1 million input tokens and $10.50 per 1 million output tokens.”
Neowin continues: The ongoing price war between OpenAI and Google, marked by recent significant price reductions from both companies, is a promising development for developers. This increased competition is expected to drive innovation, leading to even more powerful and accessible large language models in the future.”
Phew, at least the lawyers are making money.
- We spoke about Intel last week – there’s no recall on its failing chips, but there is an extended warranty
No, there’s no recall yet, but Intel has been forced to extend warranties for its chips by an extra two years.
The comapny’s 13th and 14th generation chips are crashing and blue screening Windows at inopportune times for some users, which is nothing to do with the Crowdstrike global outage of a couple of weeks ago, but the extended warranty is the closest thing to a recall yet.
Why? Because some of the chips that have received too much voltage are already damaged, with that damage being permanent. The fix that is due in the next week or two can’t fix damaged chips, but this leaves the end user needing to contact Intel – or the company they bought the computer from – to get a replacement.
So… if your computer is from the last couple of years and has been playing up – and the update that comes doesn’t fix the problem – talk to the company you bought the computer from and they’ll be able to help.
More from Tom’s Hardware here.
- It’s a bit of a weird week in tech. Now we have NVIDIA, who makes the H-100 AI chips that power ChatGPT and other AI systems, discovering a massive design flaw late in the game, affecting and delaying 10s of billions of dollars for its new B200 chip.
Well, NVIDIA has discovered these unexpected design flaws just as they were about to flick the switch to a massive production run of the processors.
Of course, it’s better to discover these flaws before hundreds of millions of these chips have been manufactured, but the reality is this will push the start date for these chips to be made by at least three months, with customers like Microsoft, Meta, Google, and others – with orders in the tens of billions of dollars – now having to wait until 2025 to get the new models, when deliveries were expected towards the end of the year.
NVIDIA gets the Taiwanese chip making giant TSMC – which makes chips for Apple, AMD and many others – to make its chips, and both companies are deep in investigations to find out what the problem is.
The replacement for the H-100 chips is called the Blackwell B-200, which are advertise as delivering 25 times more performance at the same power while reducing water consumption, so this delay is a big deal in getting smarter AI.
Despite recent worries, “it remains clear that demand levels continue to rise, with all major hyperscalers continuing to grow their capex outlooks,” Bernstein analyst Stacy Rasgons said. “Nvidia’s competitive window is so large right now that we don’t think a three-month delay will cause significant share shifts.”
Here’s more from PCMag.
- Last week we saw Google getting serious flack for running its Olympic-themed ad to promote its Gemini AI, where a father was helping his daughter use AI to write a fan letter to her favourite Olympic world-record holding runner instead of helping her write a letter from scratch and from the heart.
Google has had to pull the ad. The backlash was so great, shaping exactly the wrong instance where AI should be used, that it pulled the ad from the airwaves.
You can still see it on YouTube, but instead of teaching his daughter to write from the heart, whether there were mistakes or not, the ad suggested just using AI, which could easily have meant a stack of fan letters that all weirdly sounded similar.
No wonder there’s mistrust, at least by some, of what this AI revolution truly brings, especially when fuelled by decades of movies showing the fight between man vs the machine.
An article online pointed to NPR pop culture podcast host Linda Holmes writing, “Sit down with your kid and write the letter with them! I’m just so grossed out by the entire thing.” She continued, “Obviously there are special circumstances and people who need help, but as a general ‘look how cool, she didn’t even have to write anything herself!’ story, it SUCKS. Who wants an AI-written fan letter??”
Now of course, the output of an AI depends upon the instructions you give it, but stifling human creativity and putting it into the hands of a machine harkened to when Apple launched its iPad Pro a couple of months ago and showed a giant compacting machine squishing pianos, guitars and other musical instruments, including statues, old arcade games machines, tins of paint and more into the ultra thin iPad Pro, but the ad caused consternation with creators worldwide, and Apple pulled that ad from the airwaves too.
What is it with big companies and their missteps? At least they’re learning that, like in the movies, humans DO fight back!
Here’s more from Variety.
- Are you still using a computer with Windows XP, 7 or 8 – or even Windows 10? An cyber hack from 2018 has sprung back to life and it’s a big one affecting 70% of the world’s Windows computers that aren’t on Windows 11. What’s the story here?
The US Cybersecurity and Infrastructure Security Agency, known as CISA, of an attacked called “APT41″ and says it is “a prolific and dangerous threat actor that all users and cybersecurity practitioners should be keeping track of.
“APT41 created a tailored loader to inject a proof of concept for CVE-2018-0824, a remote code execution vulnerability in Microsoft COM for Windows, directly into memory to achieve local privilege escalation… Users should ensure all Windows systems are up to date to the latest version to protect against this vulnerability.”
Older Windows systems are NOT protected from this threat, and if you’re using Windows 10 and haven’t been updating it, you’re at risk too. Windows 11 users aren’t at risk, but it was only last month that Windows 11 finally crossed the 30% usage mark, with many more people simply not upgrading.
If you need to use older computers for access to older software or hardware, don’t connect them to the Internet, just use them offline.
If you have Windows 10 and can get the free upgrade to Windows 11, which is now three years old – then now is a great time to do so – after making backups of your data of course, just in case.
But using old computers, like old phones, smartwatches and tablets that aren’t getting updates anymore is dangerous, as these devices are actively under attack by the world’s cybercriminals, so either update your apps and operating system – or look to getting a new computer or device.
Here’s more from Forbes.
- Google has issued a warning for a new and serious zero-day attack, but what does this mean for Samsung and other Android users?
Forbes notes this bug is called an “Android kernel vulnerability” – the heart of the Android operating system – which could lead to remote code execution with system execution privileges, which means the bad guys can into your phone over the Internet, and run any code they want.
This could be a crypto miner to generate bitcoins, it could be to steal your passwords, turn own your camera and microphone, steal your photos and other data.
Google will shortly release the August security update for its Pixel smartphones and tablets – I checked and the update isn’t available yet and will take a couple of days at least – but given Google issues its updates very quickly, many other manufacturers aren’t as quick off the mark as Google.
Google warns “There are indications that CVE-2024-36971 may be under limited, targeted exploitation”, so it has a genuine need to get this fix out there quickly, even if it’s not actual panic stations just yet.
Now, Samsung has just released its own security update for its version of Android – patching 50 security vulnerabilities, so its a must-download for users – but this update doesn’t include a fix for this newly announced problem. That means Samsung users will likely have to wait until September to get a fix, but this vulnerability could be so serious that Samsung may have to act sooner and release yet another security update during August to deal with this problem.
Of course, most Android makers aren’t as studious as Samsung or Google, and their updates can be slower, although many but not all phones that are from the last three or four years are still getting updates.
If you know what you’re doing, you can check for the month and year of your last security update in the settings where you check for updates. If the security updates aren’t from 2024, your phone is ripe for replacement, but the only way to truly get the very latest updates quickly is with a Google phone, the cheapest of which brand new is double or more of the price that much cheaper units sell for.
- It will finally be easier for Aussies travelling to the US to get though passport control faster at the airport with the US Global Entry program, but for everyday travellers, they’ll have to wait until next year – so what are the details and what will this cost?
1000 Aussie tech executives who have travelled to the US five times or more in the past year will be the first to gain access to the US Global Entry program, which Australia’s Foreign Minister, Senator Penny Wong, saying the program will simplify and speed travel for the hundreds of thousands of Australians who visit the US annually.
The program offers pre-approved, low-risk travellers faster processing through US airports and quicker security clearances.
To join the program, once it is opened up to the public, you must submit an online application, pass a background check, attend a personal interview, and pay a US$100 application fee.
Otherwise you’ll just have to keep waiting in line like everyone else!
Here’s more from ABC News Australia.
