The Thought Process May Be a Problem: Microsoft and Copilot Fees
February 4, 2025
Yep, a dinobaby wrote this blog post. Replace me with a subscription service or a contract worker from Fiverr. See if I care.
Here’s a connection to consider. On one hand, we have the remarkable attack surface of Microsoft software. Think SolarWinds. Think note from the US government to fix security. Think about the flood of bug fixes to make Microsoft software secure. Think about the happy bad actors gleefully taking advantage of what is the equivalent of a piece of chocolate cake left on a picnic table in Iowa in July.
Now think about the marketing blast that kicked off the “smart software” revolution. Google flashed its weird yellow and red warning lights. Sam AI-Man began thinking in terms of trillions of dollars. Venture firms wrote checks like it was 1999 again. Even grade school students are using smart software to learn about George Washington crossing the Delaware.
And where are we? ZDNet published an interesting article which may have the immediate effect of getting some Microsoft negative vibes. But to ZDNet’s credit the write up “The Microsoft 365 Copilot Launch Was a Total Disaster.” I want to share some comments from the write up before I return to the broader notion that the “thought process” is THE Microsoft problem.
I noted this passage:
Shortly after the New Year, someone in Redmond pushed a button that raised the price of its popular (84 million paid subscribers worldwide!) Microsoft 365 product. You know, the one that used to be called Microsoft Office? Yeah, well, now the app is called Microsoft 365 Copilot, and you’re going to be paying at least 30% more for that subscription starting with your next bill.
How about this statement:
No one wants to pay for AI
Some people do, but these individuals do not seem to be the majority of computing device users. Furthermore there are some brave souls suggesting that today’s approach to AI is not improving as the costs of delivering AI continue to rise. Remember those Sam AI-Man trillions?
Microsoft is not too good with numbers either. The article walks through the pricing and cancellation functions. Here’s the key statement after explaining the failure to get the information consistent across the Microsoft empire:
It could be worse, I suppose. Just ask the French and Spanish subscribers who got a similar pop-up message telling them their price had gone from €10 a month to €13,000. (Those pesky decimals.)
Yep, details. Let’s go back to the attack surface idea. Microsoft’s corporate thought process creates problems. I think the security and Copilot examples make clear that something is amiss at Microsoft. The engineering of software and the details of that engineering are not a priority.
That is the problem. And, to me, it sure seems as though Microsoft’s worse characteristics are becoming the dominant features of the company. Furthermore, I believe that the organization cannot remediate itself. That is very concerning. Not only have users lost control, but the firm is unconsciously creating a greater set of problems for many people and organizations.
Not good. In fact, really bad.
Stephen E Arnold, February 4, 2025
Microsoft and Bob Think for Bing
February 4, 2025
Bing is not Google, but Microsoft wants its search engine to dominate queries. Microsoft Bing has a small percentage of Internet searches and in a bid to gain more traction it has copied Google’s user interface (UI). Windows Latest spills the tea over the UI copying: “Microsoft Bing Is Trying To Spoof Google UI When People Search Google.com."
Google’s UI is very distinctive with its minimalist approach. The only item on the Google UI is the query box and menus along the top and bottom of the page. Microsoft Edge is Google’s Web browser and it is programed to use Bing. In a sneaky (and genius) move, when Edge users type Google into the Bing search box they are taken to UI that is strangely Google-esque. Microsoft is trying this new UI to lower the Bing bounce rate, users who leave.
Is it an effective tactic?
“But you might wonder how effective this idea would be. Well, if you’re a tech-savvy person, you’ll probably realize what’s going on, then scroll and open Google from the link. However, this move could keep people on Bing if they just want to use a search engine. Google is the number one search engine, and there’s a large number of users who are just looking for a search engine, but they think the search engine is Google. In their mind, the two are the same. That’s because Google has become a synonym for search engines, just like Chrome is for browsers. A lot of users don’t really care what search engine they’re using, so Microsoft’s new practice, which might appear stupid to some of you, is likely very effective.”
For unobservant users and/or those who don’t care, it will work. Microsoft is also tugging on heartstrings with another tactic:
“On top of it, there’s also an interesting message underneath the Google-like search box that says “every search brings you closer to a free donation. Choose from over 2 million nonprofits. This might also convince some people to keep using Bing.”
What a generous and genius tactic! We’re not sure this is the interface everyone sees, but we love the me too approach from monopolies and alleged monopolies.
Whitney Grace, February 4, 2025
Another Bad Apple? Is It This Shipment or a Degraded Orchard?
February 3, 2025
Yep, a dinobaby wrote this blog post. Replace me with a subscription service or a contract worker from Fiverr. See if I care.
I read “Siri Is Super Dumb and Getting Dumber.” Now Siri someone told me had some tenuous connection to the Stanford Research Institute. Then the name and possibly some technology DNA wafted to Cupertino. The juicy apple sauce company produced smart software. Someone demonstrated it to me by asking Siri to call a person named “Yankelovich” by saying the name. That just did not work.
The write up explains that my experience was “dumb” and the new Apple smart software is dumber. That is remarkable. A big company and a number of mostly useful products like the estimable science fiction headset and a system demanding that I log into Facetime, iMessage, and iCloud every time I use the computer even though I don’t use these features is mostly perceived as one of the greatest companies on earth.
The write up says:
It’s just incredible how stupid Siri is about a subject matter of such popularity.
Stupid about a popular subject? Even the even more estimable Google figured out a long time ago that one could type just about any spelling of Britney Spears into the search box and the Google would spit out a nifty but superficial report about this famous person and role model for young people.
But Apple? The write up says from a really, truly objective observer of Apple:
New Siri — powered by Apple Intelligence™ with ChatGPT integration enabled — gets the answer completely but plausibly wrong, which is the worst way to get it wrong. It’s also inconsistently wrong — I tried the same question four times, and got a different answer, all of them wrong, each time. It’s a complete failure.
The write up points out:
It’s like Siri is a special-ed student permitted to take an exam with the help of a tutor who knows the correct answers, and still flunks.
Hmmm. Consistently wrong with variations of incorrectness — Do you want to log in to iCloud?
But the killer statement in the write up in my opinion is this one:
Misery loves company they say, so perhaps Apple should, as they’ve hinted since WWDC last June, partner with Google to add Gemini as another “world knowledge” partner to power — or is it weaken? — Apple Intelligence.
Several observations are warranted even though I don’t use Apple mobile devices, but I do like the ruggedness of the Mac Air laptops. (No, I don’t want to log into Apple Media Services or Facetime, thanks.) Here we go with my perceptions:
- Skip the Sam AI-Man stuff, the really macho Zuck stuff, and the Sundar & Prabhakar stuff. Go with Deepseek. (Someone in Beijing will think positively about the iPhone. Maybe?)
- Face up to the fact that Apple does reasonably good marketing. Those M1, M2, M3 chips in more flavors than the once-yummy Baskin-Robbins offered are easy for consumers to gobble up.
- Innovation is not just marketing. The company has to make what its marketers describe in words. That leap is not working in my opinion.
So where does that leave the write up, the Siri thing, and me? Free to select another vendor and consider shorting Apple stock. The orchard is dropping fruit not fit for human consumption but a few can be converted to apple sauce. That’s a potential business. AI slop, not so much.
Stephen E Arnold, February 3, 2025
A Failure Retrospective
February 3, 2025
Every year has tech failures, some of them will join the zeitgeist as cultural phenomenons like Windows Vista, Windows Me, Apple’s Pippin game console, chatbots, etc. PC Mag runs down the flops in: “Yikes: Breaking Down the 10 Biggest Tech Fails of 2024.” The list starts with Intel’s horrible year with its booted CEO, poor chip performance. It follows up with the Salt Typhoon hack that proved (not that we didn’t already know it with TikTok) China is spying on every US citizen with a focus on bigwigs.
National Public Data lost 272 million social security numbers to a hacker. That was a great day in summer for hacker, but the summer travel season became a nightmare when a CrowdStrike faulty kernel update grounded over 2700 flights and practically locked down the US borders. Microsoft’s Recall, an AI search tool that took snapshots of user activity that could be recalled later was a concern. What if passwords and other sensitive information were recorded?
The fabulous Internet Archive was hacked and taken down by a bad actor to protest the Israel-Gaza conflict. It makes us worry about preserving Internet and other important media history. Rabbit and Humane released AI-powered hardware that was supposed to be a hands free way to use a digital assistant, but they failed. JuiceBox ended software support on its EV car chargers, while Scarlett Johansson’s voice was stolen by OpenAI for its Voice Mode feature. She sued.
The worst of the worst is this:
“Days after he announced plans to acquire Twitter in 2022, Elon Musk argued that the platform needed to be “politically neutral” in order for it to “deserve public trust.” This approach, he said, “effectively means upsetting the far right and the far left equally.” In March 2024, he also pledged to not donate to either US presidential candidate, but by July, he’d changed his tune dramatically, swapping neutrality for MAGA hats. “If we want to preserve freedom and a meritocracy in America, then Trump must win,” Musk tweeted in September. He seized the @America X handle to promote Trump, donated millions to his campaign, shared doctored and misleading clips of VP Kamala Harris, and is now working closely with the president-elect on an effort to cut government spending, which is most certainly a conflict of interest given his government contracts. Some have even suggested that he become Speaker of the House since you don’t have to be a member of Congress to hold that position. The shift sent many X users to alternatives like BlueSky, Threads, and Mastodon in the days after the US election.”
It doesn’t matter what Musk’s political beliefs are. He has no right to participate in politics.
Whitney Grace, February 3, 2025
AI Smart, Humans Dumb When It Comes to Circuits
February 3, 2025
Anyone who knows much about machine learning knows we don’t really understand how AI comes to its conclusions. Nevertheless, computer scientists find algorithms do some things quite nicely. For example, ZME Science reports, "AI Designs Computer Chips We Can’t Understand—But They Work Really Well." A team from Princeton University and IIT Madras decided to flip the process of chip design. Traditionally, human engineers modify existing patterns to achieve desired results. The task is difficult and time-consuming. Instead, these researchers fed their AI the end requirements and told it to take it from there. They call this an "inverse design" method. The team says the resulting chips work great! They just don’t really know how or why. Writer Mihai Andrei explains:
"Whereas the previous method was bottom-up, the new approach is top-down. You start by thinking about what kind of properties you want and then figure out how you can do it. The researchers trained convolutional neural networks (CNNs) — a type of AI model — to understand the complex relationship between a circuit’s geometry and its electromagnetic behavior. These models can predict how a proposed design will perform, often operating on a completely different type of design than what we’re used to. … Perhaps the most exciting part is the new types of designs it came up with."
Yes, exciting. That is one word for it. Lead researcher Kaushik Sengupta notes:
"’We are coming up with structures that are complex and look randomly shaped, and when connected with circuits, they create previously unachievable performance,’ says Sengupta. The designs were unintuitive and very different than those made by the human mind. Yet, they frequently offered significant improvements."
But at what cost? We may never know. It is bad enough that health care systems already use opaque algorithms, with all their flaws, to render life-and-death decisions. Just wait until these chips we cannot understand underpin those calculations. New world, new trade-offs for a world with dumb humans.
Cynthia Murrell, February 3, 2025
-
- Subscribe to Beyond Search
Feature archive
News archive
-
