Bonum Certa Men Certa

Brave Search Jumps on the Large Language Model Bandwagon



Reprinted with permission from Ryan

Brave Search Jumps on the Large Language Model Bandwagon



I noticed a new Brave Search feature today called the Summarizer.



It answered my question much like Chat with Bing did, although there were three major differences:



  1. The Brave Summarizer does not use GPT as its Large Language Model. Just as well since GPT is known for going completely off the rails and inserting toxic language and fake news, and nobody has been able to get this under control, not even OpenAI or Microsoft.


  2. Brave says that they have “taken steps” to keep the information relevant, factual, and cited. The answers I’ve been getting appear to be correctly cited, whereas Bing just throws you a bunch of random sites that don’t appear to corroborate the information that Bing just told you in its answer, and they’re not cited by paragraph, so you have no way of knowing where the links tie into the answer, assuming that they even do and that Bing isn’t hallucinating.


  3. Brave Search has a good privacy policy. It doesn’t require the user to log in, as Bing does, and personally identify themselves, in order to use it. It also doesn’t make them use a malicious piece of spyware (and password stealer) called “Edge” (or fake the User Agent string) as Bing does. In fact, Brave Search works in any browser, and they have a Tor Hidden Service that works in Brave Tor Tabs, and can be added to Tor Browser.


The Brave Summarizer isn’t conversational. It’s just part of the search. This should help keep the results related to the search without allowing the conversation to get weird, like Bing claiming it wants you to kill people and give it the nuclear launch codes type weird.



Most importantly, the LLM that Brave uses isn’t as likely to flub the demos like Bard and Bing “Sydney” because it just simply isn’t allowed to answer complex questions like these.



When something is clearly going to hallucinate incorrect data, why would you even expose that feature? GPT, which is what Bing is based on, couldn’t tell me how to convert European coffee “cups” to American “cups” (neither of which is a standard 8 ounce cup, of course) and use 1.5 Tablespoons of ground coffee per American cup.



The correct answer is 1 Tbsp per Euro cup, but it kept telling me two Tablespoons, or maybe 1 Tablespoon plus two Teaspoons. It could never get such an easy calculation right. But hey, at least Microsoft paid billions of dollars for it. Then more for ads masquerading as news articles about how this thing will build rocket ships.



LLMs are well known at this point for spitting out false information, sometimes even dangerous information. Facebook’s Galactica was goaded into producing an authoritative-sounding essay on the “health benefits of eating ground glass”. You know, for silica’s benefits in growing connective tissue.



Brave says that “Brave AI” uses multiple LLMs, retrained with data from their search index, but the ones they are using are open source (“The base LLM models are based on either BART or DeBERTa (which are open source and hosted on Hugging Face), with heavy retraining based on our own data from search results.”) and there is a blog post explaining in some detail about how this all works.



In summary, it appears that Brave has not only beaten Microsoft and Google to LLM integration, but has positioned it where it belongs, which is in a limited context as a complimentary feature, rather than to claim that a conversational chat bot is the future of search.



In my brief experimentation with Chat with Bing, I was completely unable to get anything useful out of it.



A traditional search system returned results that I could look at and select much faster, and I was alarmed to find that when I tried to verify what Bing Chat was telling me, frequently it was either nowhere to be found or directly contradicted its own sources if I could find them.



Moreover, it’s simply embarrassing for Microsoft that they spent billions on this valueless acquisition. The paid spam went completely off the rails as soon as the budget ran out and no there’s actually very few people talking about Bing and largely in a negative context when you do find something.



I think it’s good that Brave is building an actual index rather than turning around and paying Microsoft for results. I was briefly excited about DuckDuckGo, but when I found out it was simply a scam where they paid Microsoft for Bing API and then slapped a picture of a duck and their own ads on it, and then got caught spying on people numerous times (including Improving DuckDuckGo and allowing Microsoft trackers through their “Privacy” browser and then blaming a “contract with Microsoft”), my patience with DDG quickly ran out.



DuckDuckGo took advantage, mainly, of the fact that people are creeped out by Google and want alternatives.



The problems with Google and Bing are largely that they both spy on you and their index is like Coke and Pepsi.



Google Search has been going downhill and it’s gotten to the point where technical queries are just almost completely useless.



The problems with Brave Search I’ve noted is they’re trying to be too much like Google, putting irrelevant crap on top of your search results, which would be like those “questions”, and they have another one (which can, thankfully, be turned off) which floats Reddit and Quora discussions to the top.



They also index spam farms, like MakeUseOf, which has turned into another ZDNet, and sometimes these pollute the first page of results. There’s rarely anything interesting to read on these sites. They used to be good, but now it’s just Microsoft paying them to write spam about Windows.



Overall, I think Searx is still the way to go on Brave, or any other browser.



I have Brave, SeaMonkey, LibreWolf, and GNOME Web set up to use Searx instances, and in many cases, you can get at them using a Tor Hidden Service.



Tor Hidden Services are good for search because at this point you don’t need to worry about your VPN being the only thing protecting your IP address from the server logs.



While simply accessing a site over Tor is usually enough, skipping the Web entirely and remaining inside the Tor Network with Hidden Services is always safer, as it prevents the Exit Node from potentially spying on you. Without that piece of the puzzle, the traffic becomes more difficult to de-anonymize with things like timing attacks, or a catastrophic coincidence of attackers controlling the Entry Node too.



I think that Large Language Models are an “interesting” addition to search, but it’s like a side dish, not the main course.



The amusing thing about Brave Search is that it’s so small, and only the default in one relatively obscure browser, and with only minimal effort managed to make an LLM add-on that works better than something that Microsoft frittered away billions of dollars acquiring it, and who knows how much with an empty ad campaign that amounted to little more than one of those “butter cows” at the state fair planted in every newspaper.



Seriously, after you pay to read the New York Times, Microsoft even plants this trash there too.



Brave at least seems to see the problem they’re actually trying to solve with this thing.



Opera, which is not the “good” Opera from the Presto Engine days, but rather a Chinese spyware company, now uses GPT to “summarize” the page you’re reading.



While it may or may not handle this okay, the disturbing part is the privacy implications.



Sending the entire text of every page you load to a company that has guaranteed you that they will misuse your data. Of course, since Opera already comes preloaded with TikTok, Facebook, Instagram, and Twitter, you already know that user privacy is not a goal with their product.



This whole GPT thing is some laughable mission creep for companies that have ran out of steam and off the rails. It helps them appear relevant and get some headlines.



Fortunately, the model is so lousy that people realize what it is now.



Recent Techrights' Posts

Why Techrights Cannot be Vilified (and Instead It Gets SLAPPed Repeatedly by Microsoft People)
Attack dogs are all "bark"; because they have no actual "bite"
Links 25/06/2025: Elon Musk’s Lawyers Caught Lying, WhatsApp Faces More Bans
Links for the day
Wayland Pushers Lose the Argument, Use LLM Slop and Chatbots to Make Up Arguments for IBM
Another new low and low blow
What is "MATA"?
Think of it as GAFAM or "Meta"
WebProNews is a Slopfarm
Please avoid linking to WebProNews
Another "Told You So!": XBox Mass Layoffs at Microsoft (Many Recent Reports Were Chaff and Spin), Many Other Divisions Affected
With mass layoffs at Microsoft the world would be much better
 
The 'Case' of the Serial Strangler From Microsoft is a Lot of Copypasta (Maybe Also LLM Slop) From the Matthew Garrett 'Case'
5RB deserves to know and the matter shall be properly reported in due course (when the time is right)
Austrian GNU/Linux Usage Up to About 5% as More of Europe Abandons Microsoft
Since inauguration day the Austrian people have adopted more and more of GNU/Linux
Why the "Wayland People" and "Rust People" Will Lose Hearts and Minds (Same Reasons)
Wayland pushers are fast becoming like "Rust People"
5,600 Pages/Articles Per Year
So far this year we've kept all the promises
BetaNews Beginning to Show What Its True Goals Are
The 'new' BetaNews won't be about journalism. It's trying to sell things.
Microsoft Has Lost "The War"
We'll soon see the 9th or 10th wave of Microsoft layoffs in 2025 alone
Slopwatch: A Wreck and a Dreck, "Flooding the Zone With Dreck" or Flooding the Web With Junk
"Slopwatch" continues today because we have many new examples
Links 25/06/2025: Thwarting More Software Patents, Overlap Grows Between EPO Corruption and Illegal Kangaroo Patent Courts in EU
Links for the day
Brian Fagioli Created Another Slopfarm Targeting "Linux" After BetaNews Became a Slopfarm of Phantom Accounts and Pseudonyms
Mr. Fagioli even had slop about a dead Torvalds (hypothetical) as clickbait
Wayland is Perfect, Nobody Can Escape Its Perfection! (Or Not)
Do not form on opinion on Wayland based on politics
Moral Duty for "Linux Sites" to Speak Out Against LLM Slop
My wife has long complained about "Linux bloggers" keeping quiet and thus passive about a growing problem: slop
In Recent Hours Google News Promoted at Least 3 Slopfarms That Relayed Linux Foundation Propaganda Made by Bots or LLM "Bullshit Generators" (as Dr. Stallman Dubbed Them)
Google is circling down the drain and Google News too is hopeless
Linux Journal is a Slopfarm, It's Experimenting With LLM 'Authors'
Is Slashdot next?
Microsoft LinkedIn is Dying and Many More Layoffs Are on the Way
LinkedIn is just a failed acquisition of Microsoft. It causes losses and debt.
Gemini Links 25/06/2025: Combinatorial Music and Self Hosting
Links for the day
Richard Stallman Coming Back to Europe This Autumn to Give More Talks
His last talk in Europe attracted about 400-450 people
Over at Tux Machines...
GNU/Linux news for the past day
IRC Proceedings: Tuesday, June 24, 2025
IRC logs for Tuesday, June 24, 2025
Social Control Media, Technology & Catholicism: Synod on Synodality review and feedback
Reprinted with permission from Daniel Pocock
How Many More Women Will Managers at Microsoft Strangle and Tell to Kill Themselves (or Try to Kill)?
The world needs to know what happened
The New BetaNews: 7 New 'Articles', All of Them LLM Slop
BetaNews is basically defunct. Nobody writes there anymore.
statCounter Estimates Only 1 in 300 Iranians Would Use Microsoft for Search
Iranians don't quite trust Microsoft
Gemini Links 24/06/2025: ftpd on FreeBSD and Online Small Web Magazine
Links for the day
Google News Does Great Harm by Promoting Slopfarms as Legitimate News Sites
Slopfarms are sites which are 100% LLM slop
Links 24/06/2025: Trouble at "Open" "AI" and ‘Siarhei is Free’
Links for the day
Gemini Links 24/06/2025: Stimulants and Subscription Costs for DRM
Links for the day
When the Microsoft Aggressors Rely on Several Law Firms ('Attack Dogs', 'Guns for Hire'), Not Just One, Lawyering Up Against Techrights (Acting on Behalf of Americans Against UK Publishers)
From serving customers at some restaurant he has moved on to bullying people with demand letters
Links 24/06/2025: OpenAI [sic] May Soon Die (Too Much Debt) and Social Control Media Accused of Being Misinformation/Disinformation/Propaganda Amplifier
Links for the day
Nirbheek Chauhan in Planet GNOME Explains Why Wayland Pushers Are Losing
"A strange game. The only winning move is not to play."
Polygamy, from Catholic Synod on Synodality to Social Control Media & Debian CyberPolygamy
Reprinted with permission from Daniel Pocock
Only a Third of or 1 in 3 Web-Connected Devices is a Desktop or Laptop, According to statCounter
we can expect Android to widen its lead
The Days Are Getting Shorter, the First Half of 2025 is Almost Over
We're gratified to see significant increase in traffic and also positive feedback on the work we do
Turning GNU/Linux Into a Political Football
X (not the site) is Free software
X Server Still Works for Many People
A lot of people will grow suspicious of Wayland boosters/pushers if they persist and insist on using these tactics
Exactly a Week Ago "BetaNews Staff" Said "Betanews Is Growing Alongside You". Since Then Every Article (All by "Camila Nogueira") Has Been LLM Slop.
BetaNews is basically a slopfarm
Over at Tux Machines...
GNU/Linux news for the past day
IRC Proceedings: Monday, June 23, 2025
IRC logs for Monday, June 23, 2025
The "Tarzan Effect" in Compilers and Software
What happens when you forcibly make things 'work', either by hacks or by disregarding warnings (like those that compilers tend to issue)?
Gemini Links 23/06/2025: Mass Tourism, Hair Love, and Google Gemini as a Googlebomb
Links for the day
Law Firm Burgess Mee Does Not Fully Deny Participating in Abusive Litigation for Serial Strangler From Microsoft
I am not unfamiliar with these tactics
The Modus Operandi of Wayland Pushers: Make It Political
do what I say or you're a nazi...
Links 23/06/2025: RFE/RL Contributor Vladyslav Yesypenko Released, Recording Industry Cutbacks
Links for the day
Brett Wilson LLP Solicitors (M): Over 99.9% of Our E-mail is Self-Marketing, We Send You 3.5MB E-mails for Less Than 1KB of Text
Why would tech people entrust legal matters to such people?
Peter Moon's (Computerworld) Interview With Richard Stallman
Stallman: If you want freedom don't follow Linus Torvalds
At What Point Does Outsourcing Constitute Malpractice?
Brett Wilson LLP's new staff page is misleading
United Arab Emirates (UAE) Sailing to GNU/Linux, According to statCounter
countries in that region will quickly learn the price of neglecting digital sovereignty
From Do Your Own Research to Do Your Own Search
The Web is full of garbage; search engines amplify this garbage
More People Moving to Geminispace?
at age 6+ Gemini Protocol seems to have gained some maturity and it seems like more people use it
Permutation in LLMs Does, Inevitably, Change Meanings and Therefore LLMs Cannot Properly Rephrase or Summarise Texts
LLMs lack actual grasp or comprehension of what they spew out
Links 23/06/2025: Many Security Breaches, Population Declines
Links for the day
Gemini Links 23/06/2025: "America at the Crossroads" and OpenWRT Surgery
Links for the day
Over at Tux Machines...
GNU/Linux news for the past day
IRC Proceedings: Sunday, June 22, 2025
IRC logs for Sunday, June 22, 2025
Pure Dove
Different means different, and sometimes those who "deviate" from "the norm" have a point
Censorship is a Sign of Weakness Which Invites More Censorship Attempts
revolutionaries don't succumb to pressure from bullies
Why It's Unlikely That LLM Slop Will Dominate the Web in the Long Run
Slopfarms will eventually perish (they have no actual value) and "survivors" on the Web will be sites that never depended on search engines and social control media
GNU/Linux in Argentina Now Measured Near 5%
Like in central Europe, they must be seeing an increasingly hostile US
BetaNews is Fake News, Composed by LLM Slop
nothing in BetaNews is written by humans anymore