Beyond Search

Google: No Break Up Required… Because Google Is Just Special

2026-09-12T15:15:16Z

Another dinobaby post. No AI unless it is an image. This dinobaby is not Grandma Moses, just Grandpa Arnold.

On the flight back to my home from Washington, DC, I read “Google Won’t Have to Break Up Its Ad Tech Business, Judge Brinkema Rules.” Does this mean that Google is not a monopoly? Does this mean that Google is special? Does this mean that any company can adopt the business approach of Google? I thought about these questions as I waited for the lavish mid-flight snacks delivered by an empathetic, attentive flight attendant. “Attentive” and “attendant.” I get it, and those ersatz, chemical infused biscuits were absolutely delicious.

Back to the write up. I noted this passage:

Google had previously proposed a number of competitive fixes for its ads business. It promised to make real-time bid amounts for open web display ads sold through AdX available to rival ad servers. Google also said it will deprecate its Unified Pricing Rules and allow publishers to set different price floors for individual bidders in GAM. And it promised to not use “first look” and “last look” privileges to adjust its bids for open-web display ads, although, according to Google, it stopped doing so years ago.

I quite liked this statement: “it promised.” Promises in a legal matter are obviously acceptable. Imagine a criminal in the act, captured on video by several eye witnesses, and documented with a real time message posted on X.com. When that murderer says, “I promise. I won’t kill again,” do we believe this person?

Thanks, Midjourney. Good enough.

The cited write up says:

During the trial’s remedy phase, she [the judge] pointed out that disrupting AdX or DFP might negatively impact the small publishers that currently use the latter for free. She added that any potential acquisition of AdX by another party (say, Microsoft, for example) could introduce new complications. She also suggested that requiring Google to implement behavioral fixes would be a quicker solution than mandating a breakup and enduring the yearslong appeals process that would come with it.

How does one interpret this possible ruling? Let me do a bit of airplane boredom imagining:

No one cares to understand the business system Google has put in place for online advertising. Therefore, just allow Google to keep on doing Googley things. No change to the advertising factories Google operates.
Google does a great job of explaining and reframing its business. Neither its attorneys, employees, the opposing attorneys, or most of the Google customers have any knowledge of Google’s business methods. Therefore, clear explanations are impossible. People just smile, nod, and use Google. Convenience wins.
Advertisers just want to get leads and make sales. A bit of grumpiness is part of the ad game. Pay the bill and move on.
Certain behaviors are now institutionalized. Period. Forget it.

Are these points valid? Of course not. I am delighted that American Airlines provides minimal customer care. Imagine eating something delicious and trying to think critically about Google promising to make changes. What’s worse? Airplane travel or the jumbled thinking of a legal proceeding related to the operation of a remarkable company.

I know my answer.

Stephen E Arnold, September 14, 2026

Thomson Reuters Launches Its Own LLM

2026-09-12T19:40:43Z

Thomson Reuters is the large news, content, and technology company. It uses the word “trust” on many of its outputs. They firm now has it own smart large language model called… wait for it … Thomson. “Thomson Reuters Leverages Its World-Class Data Assets To Launch Its Own Frontier Model” controls this model, and it may reduce the firm’s need to hire lawyers to do certain lawyerly things like read cases, mark them up, and generate an annotation. My hunch is that non-lawyers at Thomson will give Thomson a spin in order to create Thomson style outputs. Is this too many Thomsons? Of course not. Lord Thomson of Fleet is thrilled with the name “Thomson.”

Instead of spending billions of dollars and waiting years to build infrastructure, Thomson Reuters used an open source foundation and invested $40 million to train its LLM on the most important information-centric task Thomson employees undertake. Because Thomson Reuters used open source software, Thomson’s hands have Thomson by the scruff of the neck. The undertaking appears to have sidestepped the costs incurred by some other AI pioneers.

Here is what makes Thomson different from other frontier models:

“Thomson starts from a strong open-source foundation. What makes it different is what happens next: state-of-the-art mid-training and post-training techniques, drawing on decades of authoritative content from Westlaw, Practical Law, Checkpoint, and Reuters, with hundreds of subject matter experts integrated from the design of training objectives through to the final evaluations…The model has been trained on less than 10% of Thomson Reuters content so far, and what comes next is not simply feeding it more data. It is continued discovery of new kinds of specialization and understanding, made possible only by building on decades of proprietary content and editorial expertise.”

Is Thomson a leader or a follower in AI? What will be Reed Elsevier’s response? What will the lawyers now employed to do the old-fashioned mark up react to having Thomson a click away on their computer screen? Will Thomson output information that makes the occasional error (just like real lawyers)?

None of these issues are important in the grand scheme of AI deployment. What’s important is [a] cost cutting, [b] efficiency, [c] marketing, and [d] showing subscribers that Thomson Reuters is definitely moving down the information highway with AI as a copilot (yeah, lame pun, but I will blame my dinobaby boss).

Whitney Grace, September 14, 2026

Public Service Post: Disabling AI

2026-09-12T18:46:43Z

Hey Big Tech and other AI lovers, did you know there are people out there who don’t want AI on their phones or devices? Yes, they do exist. There are actually a lot of them. Many of these users do nothing about the AI on their phone and just move on with their lives. The proactive ones do something about it. How do you do however? Librarian.net posted an article called, “How To Disable Or Avoid Intrusive AI”:

“One of the biggest questions I get at Drop-In Time at the library (besides “what is taking up all my cloud storage?”) is how to disable or avoid intrusive AI that shows up where people don’t want it. This is a guide for people who would like less intrusive AI in their tech environment. Maybe you like AI and find it useful? That’s fine, this document is probably not for you. Additions/edits welcome. Leave a comment or drop me an email.”

The article covers how to disable the AI capabilities on Adobe Acrobat and Adobe Reader. There are different guides for Android and Apple, so users of different mobile device will have specific information for them.

There is also a guide on how to disable AI in browsers too. If you want to disable AI in other applications the guide includes: WhatsApp, Zoom, Yahoo Mail, Slack, Windows, and Google Workspace.

This is a terrific guide for turning off the AI. Now if only more people would be willing to turn off AI and send Big Tech a message.

Whitney Grace, September 14, 2026

The Chainsaw Tesla Dude: Another Play Fails

2026-09-07T16:59:01Z

Here’s some news about the infamous Grokipedia that is run and edited by Elon Musk’s AI agent Grok. Law Fare Media reports that, “Grokipedia Stopped Reviewing Edits In April. It Didn’t Tell Anyone.” If you were unaware, Grokipedia is an AI written alternative to Wikipedia. It’s entries are maintained by a large language model that decides what to write and when it will edit itself. The AI encyclopedia hasn’t been updated since April 24. When this article’s authors sent updates to Grokipedia, they were not taken and attributed by the AI.

This is important because:

“As agentic AI systems begin to generate and maintain reference knowledge independently, the question is not only whether their output is accurate at a given moment. It’s whether their behavior over time is observable, and their record stable enough to be accountable for their contributions to the public information ecosystem. At the moment, Grokipedia is neither.”

TechDirt reported on the same issue in, “Elon Musk’s Grokipedia Quietly Stopped Updating. Basically No One Noticed.” Unlike the previous article which delves into the statistics and web history off Grokipedia this article reports more on why this happened. It also quotes a lot of the Law Fare Media article as well, but it leaves us with this thought:

“It’s entirely possible that someone at xAI (or SpaceX or wherever the checks get cut these days) will flip the server back on and get things started again, but given how much the media hyped up Grokipedia when it launched as a potential “Wikipedia killer,” shouldn’t at least some of them acknowledge what a total failure it has been?”

It makes you realize that AI editing it great until it just quits or is told to quit. The real reason it was shut down was probably due to money because it got too expensive to maintain. This will definitely be noted as a Musk failure on his Wikipedia page.

Net net: The full self driving of Wikipedia clones. Is that a good tag line?

Whitney Grace, September 11, 2026

What Is the Future of Surveillance in the EU?

2026-09-07T15:16:27Z

Answer: Bright.

We have another great idea from the European Union in regards of a Big Brother state says Reclaim The Net in the story: “European Commission Revives Push For Encryption Backdoors In Protection Strategy, Framing Mass Surveillance As ‘Lawful Access.’” Can anyone else here a collection groan coming from Europe and other continents about this policy? This new policy is called ProtectEU and it would undermine end-to-end encryption in order to give law enforcement more power. There is a vision and a workplace but no proposals have been set in stone. The cited article offers:

“A press release asserts that the current geopolitical environment is one of “growing” threats from hostile states, and mentions powerful criminal groups and terrorists who are “operating increasingly online” – as well as “surging cybercrime and attacks against our critical infrastructure.” With the threat elements defined in this way, the EU’s new strategy focuses on six areas, one of them being “more effective tools for law enforcement” – which is where online encryption comes under attack. When it describes how the groundwork might be laid for mandating encryption backdoors, the EU chooses to use euphemisms such as creating roadmaps for “lawful and effective access to data for law enforcement” and seeking “technological solutions for accessing encrypted data.”

If a technology roadmap were established, these solutions would be implemented on a known cadence. Civil rights organizations, tech companies, and privacy advocates are lobbying against policies that would ban encryption.

A more interesting question to consider is, “How likely is it that mass surveillance will become the normative behavior in the US for governmental and commercial entities?” Commercial surveillance is in place in my opinion. The US government is surprised each time a story reveals that data brokers can provide access to information that reveals the location of certain government facilities and staff.

The US government itself has departments interested in mass surveillance. However, the US government’s ability to organize, deploy, and make use of technology seems to be a work in progress. My view is that the US will watch what Europe does and then “test” features and functions that make sense in America.

What does this mean for future Americans? Hey, good question. Why don’t you answer it in a blistering exposé of the current Administration. Let me know how that works out for you, please.

Whitney Grace, September 11, 2026

Are Bad Actors Upping the Ante

2026-09-07T16:59:42Z

Remember the good old days when a bad actor would scam a grandma down the block out of $40,000? Those days seem to be morphing into larger scale actions.

Bad actors are holding a German city hostage says the BBC in the article, “Berlin Is Being Blackmailed By Hackers, Mayor Says.” The black hat hackers stole data and compromised the city’s systems. The bad actors demanded a ransom be paid in order to release the city, but Berlin officials refused to pay. Berlin officials were forced to close down the city’s online systems and investigated what data had been stolen.

The Rhysida group took responsibility for the attack. These bad actors are believed to be located in Russia or eastern Europe. The Rhysida group previously attacked the British Museum. It is believed they are demanding 30 bitcoin €2m (£1.7m) as the ransom amount.

The Beeb reported:

“It has said on its website it intends to begin auctioning the 5.79 terabytes of data it was able to steal from Berlin in seven days’ time, according to news agency Reuters. “Berlin will not be blackmailed,” Wegner said on Friday. He said state police, prosecutors and federal security services were working to investigate the suspected perpetrators “with the utmost urgency”, adding that inquiries into the “content and scope of the compromised data are being pursued with great intensity”. Officials for the city-state said an initial data leak occurred between 7 and 12 August. Then, on 14 August, two department networks were shut down, making applications for housing benefits and payments impossible for several days. Forensic investigations had since revealed further data leaks within Berlin’s transport and environment department, officials said.”

The Rhysidia group published a countdown on their website that indicates when they will auction off Berlin’s data. We’re looking at the future of cyber crime. Forget grannies. Go big time seems to be looming in the future. (Is AI responsible? Of course not.)

Whitney Grace, September 11, 2026

Blindsided by the Google: Get Used to It, Pal

2026-09-07T14:10:53Z

The person who wrote “Google Dropped Two-Thirds of Our Site from Its Index in a Single Day. We Found the Culprit — Except It Wasn’t Him” is a diligent worker. His collection of tools, his measured response to getting Googled in the nether region, and his genuine puzzlement allow his technical carbon fiber to glint.

Please, navigate to the source article and work through the processes the article’s author used to try and figure out why he has been walking around very carefully since that sudden Google insertion. After several onerous attempts to figure out how his “help other people” Web site went from hero to zero, the final conclusion was, “No one knows.”

The SEO wonks will definitely sell some magic prune juice to the author of the article. My hunch is that the Google insertion will remain lodged. Furthermore, my experience suggests that nothing short of creating a new site and rolling out the content in cadenced chunks will fix the problem. Buying some Google ads to point at the new site will help a tiny bit.

Is it possible that the Google has an undocumented dependency? The author of the article stumbled into an animal trap and has been impaled by a Google programmer’s carelessness or ineptitude.

Net net: This article makes clear that Mother Google sometimes wanders around in a cloud of unknowing accompanied by sprinkles of indifference and arrogance.

Stephen E Arnold, September 10, 2026

India Wants Deep Fake Videos Labeled

2026-09-07T14:47:34Z

Nice idea, but that may be impossible with the current technology.

In the US BAITs (big AI tech companies) may have the US government firmly in their grasp, but other countries are saying “not so fast, bro”. One of those apparently retrograde countries is India. The Tech Times story “Indian Court Forces Meta, Google To Reveal Who Made Deepfakes Of Cabinet Minister” states: India’s Bombay High Court issued a interim earlier in August that directed Google and Meta to delete AI-generated deepfake videos “falsely linking Union Minister for Road Transport and Highways Nitin Gadkari to corruption around the country’s E20 ethanol-blending fuel program.” They also demand that the anonymous identity data of the account holders be handed over to the government. This is? ? groundbreaking development because India is a country that doesn’t have any standalone law relating to the criminalization of deepfakes that attack politicians.

This appears to be big deal in India:

“The ruling is the first known instance of an Indian court applying the personality-rights deepfake framework — previously used exclusively to protect the commercial likenesses of film and television celebrities — to a sitting member of the national cabinet. That extension matters beyond this specific minister: every Indian politician facing an AI-generated disinformation campaign now has a legal template for forcing platforms to reveal the people behind it. MediaNama’s analysis of the suit filing identified this as a significant structural shift in how Indian courts approach deepfake enforcement.”

India is the last country you would expect to take action on deepfakes given that it is still a developing country. In other words, they are a mess. Yes, the United States is a mess, but not as bad as India. However, the United States has something to learn from India as well as the European Union when it comes to AI. India may prove to be as much a problem for Big Tech as the EU. We say awesome.

Whitney Grace, September 10, 2026

Humans Matter: An AI Game

2026-09-07T14:33:33Z

The Scale X blog talks about a fun browser game where users play the human-in-the-loop for an AI coding agent that approves or denies command under a clock. Some of the commands are routine, while others indicate that the AI algorithm has been compromised and is sending information to a remote server.

The game developer kept stat of how the players did in the human-in-the-loop situation and discovered that 7% of the players approved every single prompt. Uh oh! Another 35.2% caught every threat, but only with 20.8% of that percentage managed to do that while blocking every one in five commands. The average player missed one in three threats and 32.9% of the sessions ended in a negative score.

This might be a game but it relates well to real world scenarios:

“Big caveat: this was a short game, where approximately 34% of the commands players saw were threats. In our day-to-day work these threats (fortunately, so far) appear rarely. Players were under artificial time pressure, but I’d argue that’s not too different in the real world where similar pressure to ship quickly may also lead to same rushed approval habits. That said I do believe the numbers below are still useful for garnering some insights on comparing threats and the general awareness of them against one another and general issue of relying on ‘human-in-the-loop’ as a valid safeguard.”

There’s also real life takeaways from this too. There is too much noise for the humans-in-the-loop show it makes them very fatigued. The tools for this type of work also need to be simpler, such as sandboxing and strict context isolation. AI algorithms are also only granted permissions when these safeguards are in place.

We know of one firm that views its AI plays as a video game. Greedy algorithms must win. Humans lose. Just a thought.

Whitney Grace, September 10, 2026

OpenAI Achieves Maximum AI Hype. Winner!

2026-09-07T14:10:09Z

The race is over. It is official. I know that OpenAI, Sam AI-Man, and the uncontrollable (ha ha) smart software is now deliveered science fiction as a service or SFAS. Gizmodo’s article “OpenAI Claims We’re in the ‘AGI Era’ With Release of GPT-6 Astra” makes this clear. Ignore the fact that the Ed Zitrons of the world like to poke fun with mock outrage at the financial set up of OpenAI. Just look the other way when the system just delivers some of those buried Swiss underwear data or goes offline due to network magic. OpenAI has turned zeros and ones into … artificial intelligence.

The write up states in a weird he said, someone said way:

According to Wired, OpenAI President Greg Brockman said, “It’s not unreasonable to feel that we are now in the AGI era,” regarding the release. “I think that if we fast-forward a couple of years, when we look back and say, ‘When was it really that AGI was created?’ I think it’s going to be about this time, and I think it might be about this model.”

I interpret this OpenAI statement as actual factual. I translated it to “Winner!” In my opinion, whether true or not, other AI outfits will be sipping their tea, lattes, and filtered infused water discussing this OpenAI statement of its reality. US companies will lose their marketing demons to reposition the OpenAI statement and point out that OpenAI is manifesting distressing psychological behaviors suggestive of profound mental disordeer. Chinese outfits (yes, I know you are there ByteDance) will just release stealth free equivalents. A few days later the beyond OpenAI models will become available at bargain basement prices. Other AI companies will point to the strengths of their narrow scope models. I am waiting to hear from you Thomson Reuters.

The article makes this statement:

Altman even said in July that he believes “we are now, like, in the singularity.” So OpenAI not quite declaring AGI but getting about as close to the line as you can without going over is in keeping with the clear effort to ramp up the hype.

Yes, Sam AI-Man, you are the point humanoid. You are a winner. Maybe you are THE winner?

Engadget reminds me:

OpenAI also made a special point to emphasize that GPT-6 is its safest model yet, with more work put into Astra’s alignment and lots of oversight of its chain-of-thought reasoning to make sure that it doesn’t go off the rails. Understandably, OpenAI would really want to ease concerns there, given that it’s still reeling a bit from the fact that one of its models (not GPT-6 but another, secret model) broke containment during training and hacked Hugging Face, and, reportedly, no one on the OpenAI team noticed it happened for about a week.

Yep, super intelligent stuff. Software created by humans so that humans cannot know what the smart software is doing. Is there any doubt that these AI wizards are the most loved, respected, and adored innovators on earth? Of course not. But now there is just one winner. How will those who perceive themselves as super human AI players react? Will they just roll over? Will they redefine smart software so they are winners in their logic space?

One thing is 100 percent certain: The love and affection of the American people for BAIT (big AI tech) outfits is only going to increase. Data centers? No problem, Sam AI-Man. Water and electricity price increases and brown outs? Hey, no big deal, right? Mass roll out of AI-centric ways to reduce personnel costs? Definitely a winner.

Yep, ladies and gentleman, we have a winner.

Stephen E Arnold, Septmber 9, 2026