Q&A: There is No Algorithm for Truth - with Tom Scott

How do you avoid bias when building an online platform? Is radicalisation driven by echo chambers or places with no regulations on free speech? Tom Scott answers audience questions following his talk.
Watch the talk: • There is No Algorithm ...
Tom Scott is a British entertainer, educator, KZbinr, web developer and former presenter of 'Gadget Geeks' on Sky One. He graduated from the University of York with a degree in linguistics. He has a popular KZbin channel with over 1.6 million subscribers and more than 325 million video views as of June 2019.
This talk and Q&A was filmed in the Ri on 27 September 2019.
---
A very special thank you to our Patreon supporters who help make these videos happen, especially:
Andrew McGhee, Dave Ostler, David Lindo, David Schick, Erik Shepherd, Greg Nagel, Jan Bannister, Joe Godenzi, John C. Vesey, Kellas Lowery, Lasse T. Stendan, Lester Su, Osian Gwyn Williams, Paul Brown, Radu Tizu, Rebecca Pan, Robert Hillier, Robert Reinecke and Roger Baker.
---
The Ri is on Patreon: / theroyalinstitution
and Twitter: / ri_science
and Facebook: / royalinstitution
and Tumblr: / ri-science
Our editorial policy: www.rigb.org/home/editorial-po...
Subscribe for the latest science videos: bit.ly/RiNewsletter

[@TomScottGo]

I wish my answers were a bit less hedged here -- but nevertheless, thank you to the lovely audience at the RI!

[@schr4nz]

You've completed an Ri presentation mate, you've made it!

[@333STONE]

You did as well as you could given the parameters. It's a tricky thing, ways to Truth, I mean. Censorship of any kind is detrimental to free society. Yet are we truly free? Thanks i enjoyed your insight

[@mrtausif1863]

Those who built the web are also not free from censorship. .😂😂

[@AkiSan0]

i felt they were good answers given the circumstances.

[@1stGruhn]

Another major issue is the reality that consensus isn't a truth maker. Who defines truth and can we know it? Modern science is built on the presumption of methodological natural materialism. Yet theism (be it pantheism or monotheism or simply dualism) has been and presently is by a large margin the dominant human viewpoint of our species history: atheism has been the minor view at all points in history. Its well documented that sociologists point to the adaptive benefits of theism: hope in hopeless situations, assurance of a 'greater' purpose in light of our present suffering/catastrophes, peace in tumultuous times since this life isn't our end, etc. But if these things explain away theism as a mere result of natural processes, yet the 'consensus' by popular belief is that its true but is factually false, how can we trust they same system to produce true beliefs. Thus we have another issue here, both that consensus isn't a truth maker and neither is adaptive benefit. Knowing reality is far more difficult than doing 'science' for we will never get away from metaphysics. Sure experiments may falsify certain aspects but proofs are something entirely different. No machine built by or influenced by man will ever nor can ever see without bias. This is why Nietzsche said might makes right: if you can force others to assent then you control the narrative and thus can dictate the consensus. This lies at the heart of all tyranny. No human or group of such is capable of being master... tyranny is the inevitable end yet without governance chaos ensues. True liberty lies in being self governed. Yet without agreement on social norms or interpersonal interaction norms absolute liberty leads to anarchy. The acceptable line of loss of freedom to what degree of tyranny is and every will be a grey area.

[@AndrewFullerton]

I really respect the fact that Tom can and will say something like "I don't know" or "I'm sure the information is available, I'm just not aware of it". Thank you for practicing what you preach and keeping things truthful!

[@evandavis5223]

Or "There are people much better qualified to talk about this than me."

[@mmmk6322]

We need more Tom's in politics... I hate the fact that politicians never say "I don't know"

[@alveolate]

leave it to the linguist to be accurate about information transmitted.

[@mattwinward3168]

But at the end there, when everyone claps, are they thanking Tom? Or are they thanking Tom's Work?

[@menachemsalomon]

They're thanking Tom, and appreciating his work. Or rather, the other way around: They're thanking Tom for the work he's done, which includes the research _and_ the presentation.

[@hashishishin]

The claps are only a parasocial bond builder. Don't question, just watch the ads ;)

[@Ludifant]

They are thanking their perception of Tom for presenting their perception of Tom´s work. He was (generally) coinciding with their preconceived views enough to be expressing gratitude in the manner expected of them by tradition and they felt pleasantly challenged and educated and probably thought they spent their afternoon wisely. And then mostly forgot about it within half an hour.

[@57thorns]

@@Ludifant That was a good description. I agree that Tom Scott rarely challenges my world view, but he does expand on it. That he is not a one trick pony, but can also be very entertaining is not a bad thing however. In my opinion Tom balances the educational side and the personal side very well. Rarely going for the obvious sponsorships or begging that is some of the KZbin/patreon/gameplay parasocial activity. Tom has built a reputation in key areas, and he is paid for giving talks in areas he has knowledge, and the occasional entertainment as well. As he should be.

[@aurelia8028]

*Woosh*

[@imaytag]

The danger of deep fakes as I see it is not that it will be easier to convince people of lies. But rather that it will be more difficult to convince people of the truth.

[@peacewalker3344]

different sides of the same coin.

[@Littleprinceleon]

To get considerable effect out of a deep fake video one has to imitate someone with broad influence: the affected persons will have to make extra effort to debunk such falsities via more reliable, more official sources of information. Since the misuse of someone's identity is one of the more serious crimes and the media consequences can be relatively quickly counteracted, the effort that goes into a deepfake versus its payoff makes such attempts in my opinion not so attractive. I think the majority of us in the "West", regardless of our other values, agrees upon the malevolence of faking others' identity. Simply because we realize how quickly it would lead to unmanageable chaos even in everyday life. As we could sadly see during the pandemic, the real problems (which last from at least the times of Sumer) simply stem from the complexity of topics involving whole societies. Questions barely exhausting even the capabilities of advanced institutions naturally stir up primal insecurities in most of us, which are then ventilated according to our personalities/worldviews (which are of course affected by actual happenings and their interpretations... So we would have to choose wisely in which activities we need/wish to engage in). Biggest long-term dangers aside economic/ecologic crises lie in unreasonable/unrealistic demands of masses on politicians, which can lead to the resurrection of such antisystemic forces which aren't able to provide viable alternatives to those financial/power/control structures they wish to replace/repair. The solution to most challenges humanity faces lies in making lots of large scale compromises (changes!) while preserving the benefits of modern ways of life, which is possible only trough innovations and some restraint. Most of us in the west would like to have life standards we are used to nowadays. However, will preservation only without much chance for improvement be motivating enough for us to stay in race with non-western lifestyle? Or will the desire for comfort and insurances "infect" such percentage of the economically more relevant population that a global collapse is inevitable?

[@Sage-vd3fg]

Tom in the full talk: Mostly serious, mildly indimidating, science talk Tom in the Q&A: A small man sitting on a huge ass desk and talking

[@JohnnyBooi]

I was wondering whether indimidating was another actual word or was it a spelling typo of intimidating because it strangely felt like such a word could exist, but I think it was a typo since I didn't find any results on google. It messed with my brain for quite a bit.

[@VivekYadav-ds8oz]

@@JohnnyBooi indimidating actually does sound intimidating, unlike "intimidating".

[@potmki6601]

@@VivekYadav-ds8oz ikr. As a not-native English speaker, I spent some time just assuming that « intimidating » had related meaning to « intimate »

[@potmki6601]

It’s somewhat confusing to see Tom Scott described as intimidating. I don’t know the guy obvs, just don’t get remotely the same impression

[@henwoda]

@@potmki6601 huh, I see it more connected to 'timid'

[@BeCurieUs]

Jesus, these are some great audience questions!

[@isaacorellana1754]

Absolutely underappreciated.

[@lowstrife]

The one about echo chamber vs. unmoderated open forum was crazy good. I did basically the exact same thing Tom did when he realized that and just went "woah boy"

[@ledzep331]

The algorithm got it right.. Tom Scott in a RI vid about algorithms..Clickbait for my social group...

[@Ludifant]

That would make us members of the same social group. Hello! See you at the meeting.

[@zealot9262]

Matt Parker tosses coins and spills water on Faraday's desk and now Tom is sitting on it? The worlds gone mad! Love you really Tom

[@TheRoyalInstitution]

A desk is made to be used! This is the third copy of the original Faraday desk and each of them has seen chemical splashes, explosions and many famous bums.

[@liliwheeler2204]

I like to think it's what Faraday would want ❤

[@ancbi]

Did I ever see someone sit on that table before? I don't think so. But I forgive him, must be draining enough to go without red T-shirt.

[@ignorasmus]

It was a mild shock for me too. I would not dare sit on that historic table. But then, I am not likely to ever even be allowed into that room...

[@dinowibisono99]

I'm pretty new here. What history?

[@ppppppqqqppp]

@@ignorasmus I mean, historical objects exist to be used. Treating it with reverence and superstition, instead of just using it as furniture with history, is honestly kind of weird.

[@ignorasmus]

@@ppppppqqqppp I see your point. (Except that I don't see any superstitious connotations here.) A lot of people - specially collectors, museum curators, enthusiasts etc might disagree.

[@ignorasmus]

@@dinowibisono99 Unless I am mistaken, it is the historic and iconic table upon which Michael Faraday himself had made his demonstrations about electromagnetism and other things.

[@fredhubbard7210]

One of the problems with KZbin is that they actually punish producers that make satisfying material. This was a great video, and now I want to do other things... if I turn it off, it will make it less likely for KZbin to recommend others like it when I return.

[@Tresquall]

Here's hoping that by commenting, KZbin notes your engagement and recommends similar stuff later.

[@symphonymelody9235]

“I don’t have the data to know weather that’s a good idea or not. I think the data is probably out there ,I’m just not familiar with it” - amazing, most people I see on KZbin will probably just spat out their opinion regardless because they feel that’s what people want, if they get asked a question then the audience expects an answer. They want so blindly to please the audience that they forget to think about what effect it will have. To be clear, I don’t think they do it out of malicious intent, they simply do what they think the audience wants. Tom (or rather Tom’s work) is a gem on KZbin and I really hope he keeps it up for years to come

[@Android480]

Good public speaking skills are impressive to me

[@howdyhamster]

"If there is a user ID attached to a user, a discussion tends to become a criticizing game. On the other hand, under the anonymous system, even though your opinion/information is criticized, you don't know with whom to be upset. Also with a user ID, those who participate in the site for a long time tend to have authority, and it becomes difficult for a user to disagree with them. Under a perfectly anonymous system, 'it's boring', you can say, if it is actually boring. All information is treated equally; only an accurate argument will work." --Hiroyuki Nishimura

[@Torchedini]

But on the other end you have that fact checking and arguing every statement you make in comments so that the message gets taken at face value is work. Frankly most people are lazy, and don't put that much effort into it. So a persons name becomes the fact check. (oh its this and this, he hasn't lied yet) Does that work? Yes and no.

[@sbs5130]

Indeed it has been when Tom exhibits his inordinate depth of thought that I find his work most interesting and inspiring.

[@adrianvasian]

you guys need to turn the audio up when exporting the footage because it's set too low. ;)

[@MiChAeLoKGB]

That reminded me to turn my speakers volume back down, before blasting something else way too loud. THX :P

[@vinnyfromvenus8188]

Just turn up the volume, that's what the volume control is meant for

[@shmotten]

@@MiChAeLoKGB THX as in 'thanks' or as in the THX sound?

[@MiChAeLoKGB]

@@shmotten it was meant as thanks :)

[@shmotten]

@@MiChAeLoKGB Haha alright, though THX would've worked just as well!

[@zakintha]

This just might be the best, and most most troubling, RI talk I've watched to date. You paint a frightening picture, but one it's better to be aware of. Thank-you.

[@bartsola8349]

3:56 love how he's talking about tiktok like this brand new, never-seen-before little app.

[@rosslmccallum]

This really nicely covers how the spread of information and authority works online. Felt like I learnt a lot! As a content creator with an audience online I have to be careful to not create unhealthy "parasocial relationships". This means focusing on helping people become a fan of my work more than just a fan of me.

[@bbgun061]

It's clear to me that *ideological* diversity is far more important to reduce bias than ethnic or gender diversity.

[@auberry8613]

Diversity of identity is needed as lived experience is important. Two white guys talking about what black guys experience means a lot less that than a black guy giving his first hand account. You also can't represent both sides of every argument in perfect equality, good example being climate change. One of those sides is almost definitely wrong and to entertain that is just to invite damaging discourse which can, has, and will further damage our society.

[@bbgun061]

@@auberry8613 discourse can never be 'damaging to society' only actions can.

[@Erikaaaaaaaaaaaaa]

@@bbgun061 The former leads to the latter. Neo-Nazis don't go and run over people with cars just because they feel like it, they do it because the political ideology behind them.

[@Leo0718]

@@bbgun061 Discourse is action.

[@RussellSnow]

Maybe narrative diversity would be better. A representative from each echo chamber. If your echo chamber is the standard leftie techie globalist one, of course you are in favor of an algorithm that can automatically purge voices that disagree with you. But there are enormous philosophical challenges in saying this or that thing is true. I would have enjoyed more discussion of why his tribe feels it has the right to arbitrate truth. The dominant narrative is not usually completely true. He made the example in the introduction to the other video, but then adopted a, "we all know who needs suppressing so I don't need to discuss what truth is" attitude.

[@jarnMod]

And then, with that Brexit example, Scott ended the talk. Now that's a nice ending, surely UK felt the chill.

[@helloofthebeach]

That's the mark of the true scientist, no matter their background: Being honest about what they don't know and curious about the answers they don't have. Neil DeGrasse Tyson could learn a thing or two.

[@jamesl8640]

7:55 beautiful question. And an equally good answer from Tom there.

[@leo_warren]

Well done Tom, Great Discourse!

[@whomst5141]

Tom would be great as a college/university professor

[@midimusicforever]

Thank you for bringing me here, algorithm!

[@Charls93xx]

This talk and the Q&A session has been utterly amazing, I literally devoured it for the past hour.

[@xomvoid_akaluchiru_987]

Thank you for the video, Ri.

[@P_Mann]

I'm interested in how we get beyond using proxy traits -- race, gender, ethnicity, etc. -- so we can achieve true diversity of perspectives. Regardless of such obvious traits, groups of tech folks who all live in the Bay Area, went to Stanford or Berkeley, and exist in a bubble of similar history and age aren't likely to be all that diverse.

[@agnosticdeity4687]

This comment is for the algorithm... ...is something I often say to support content producers that I like.

[@bwcbiz]

A great presentation and Q&A. There are so many pit-falls in "the algorithm" that a big worry for me is consolidation. If there are dozens of platforms, each with their own moderation, we are better protected than if we are limited to 2 or 3 big platforms with advertiser-moderated content. You discussed 'advertiser moderation' in the sense of the algorithm trying to maximize revenue, but we also see advertiser/investor moderation when the advertisers pull ads due to allegedly offensive content or the investors (say, China) impose their own censorship based on their ownership of the platform. With the global reach of a small number of platforms, we could end up with a least common denominator of inoffensiveness that goes against every western value of free speech, rather than the echo chamber or noise overpowering information scenarios.

[@barefootalien]

I... don't know if you've paid attention to the internet much, but... gravitating toward oppressive _inoffensiveness_ is... _really_ not a concern. xD

[@crazydragy4233]

​​@@barefootalien You truly misunderstood their comment if you say that. They make it clear that a place like China dictating inofensiveness enforces cultural norms people in the west would argue are barbaric. It's literally already and issue and was brought up in this talk by Tom When different people existing, when different ways of life, historical events and science are deemed offensive because they challenge the stories cultures and governments tell, oppressive inoffensiveness dictated by the majority is a nightmare.

[@marcorito]

As always, such great insight!

[@orkstuff5635]

Good talk, unrestricted freedom of speech or absolute censorship - set your slider.

[@mito._]

I think the missing common denominator here is common sense.

[@gro_skunk]

Easy solution, tell the algorithms to reduce the cuntiness in everyones lives

[@Ludifant]

@@mito._ No such thing. It´s only common sense, that there is such a thing as common sense, making it a closed loop without meaning.

[@Momo14198]

@@mito._ "common sense". Are you serious?

[@jerryhu9005]

He also mentions that they both give rise to a similar result (radicalization of viewpoints, exodus of moderate participants) So the slider is shaped like a horseshoe with the ends very close together

[@fat_vegan_slim]

Watching this 4 years later, and Tom Scott explaining what the clock app is is very odd. It's so well known now that I didn't even name it and you still know what I mean!

[@michaelhayes7849]

We need an the 10 debate standards/rules used in the programs.

[@DasGrinch]

Facebook still lets people report your name as fake, and they require a government ID to be sent to them in order to prove it is you. One of my friends got hit last week, and the name she used on the platform for at least the last 6 years I've known her was taken away.

[@Peter-tp9nv]

Respect .......

[@sgordon8123]

Tom great talk thanks. *** Your observation about how groups become extreme over time online applies to real world institutions as well. You get a culture... each school is fairly unique like that. And we see it in police forces etc Machiavellian politics plays out all over the place in petty office politics too. Humans are a troublesome species.

[@davidioanhedges]

The algorithm isn't working - Tom Scott, who I follow, doing a Q&A after a talk at the RI ....took 4 years to be recommended to me ...

[@hvanmegen]

**applauding**

[@MrPSyman3]

That's why I like Scott. He's the only public guy I know that never reaches a conclusion and never has an answer. He simply shotguns all the variables and let's us decide. He may well and probably does have an opinion of his own but he never shares it with the viewer. Too many people think they know the answer and get betrayed by their own knowledge and expectation

[@uwuvision3211]

Oh my goodness, this reminds me of the exhibition in the Barbican, "Apple to Anomaly", where it's lots and lots of pictures from a search engine that are "related pictures". From a single picture of an apple, the search engine recommended pictures with key words until it got to politically charged words like "Traitor" and pictures related. It was so so interesting (and one of the words was "rat" so it was just cute pictures pf rats).

[@shexdensmore]

7:07 The USA had or has an Affirmative action law or programs when I was a child in the 90's. I think it started in the early 70's or 80's So there should be a lot data on it.

[@MrSonny6155]

Humanity is an echo chamber for humanity. But at least it represents all relevant parties as humans. But big topics in small groups? That is where an echo chamber becomes an echo chamber, for better or for worse.

[@grim66]

I think Tom's response to the radicalization question lines up pretty closely with what I was thinking, myself -- It's easier for an echo chamber to radicalize _the people in it_, but an echo chamber is hard to fall into from the outside in. A free-speech zone is less likely, in theory, to radicalize people on its own… But it makes it VERY easy for people to get lured into the echo chambers.

[@you2tooyou2too]

re: 5:50 When ever someone says, "draw that line" they almost inevitably should have said, "there is no line in this gray area". The speaker is addressing a (perhaps the) fundamental problem of life, "how to discern the truth from noise & lies".

[@raynscloud8072]

Holy smokes but, when he's listening to questions instead of speaking, Tom absolutely looks like Brent Spiner (the actor who portrayed Data) back in the days of ST:TNG, especially around the 13:00 minute mark. He even has that side smirk (smirk is a bad word choice, but I don't know what else to call it) that Spiner has.

[@elisam.r.9960]

Grin, perhaps?

[@macbitz]

Great lecture. So these social platforms are enabling and amplifying what appears to be basic human nature (hate, intolerance, deceit etc.), so maybe the focus should be on addressing these basic bahaviours rather than the algorithms?

[@manuxx3543]

The risk is that people or institution go for the easy way of adressing that Unification of thought/opinion, rather than trying to build brigde between each and every possible thought and opinon

[@manuxx3543]

But yeah school or something else should definitly teach us about our cognitif bias, critiqual thinking, and how to communicate and understand others. I think it should be like language, you use it in everything and it's a big domain Well those thing not teached yet should be just as important and big For now it's just up to the parents to try and teach some of them

[@cedleonard123]

Can't change human nature. I fear it may be a problem that is much too tremendous to really solve. I think the best thing we can do is to promote peaceful reasoned but honest discourse. I believe near-absolute free speech (with the right to mute) must prevail because it seems like the most straightforward way to make it so people are not be at each other's throats. Whatever "hate and intolerance" mean, they're better than violence. Discourse, self-restraint and the practice of ethics are some of the key things that separate us from the beasts. But using that as a moral foundation presumes that people will be thinking reasonably and speaking in good faith. For sure that raises doubts. The answer to these doubts is responsibility. If we spread the word (giving arguments that everyone must be responsible for their own actions), fewer and fewer people will be able to come up with excuses for doing evil. To the extent that this might be possible but not quite so, this is the only hole I can see in my view, but at least it'll mean that whenever we're practising morality we'll always be working on fixing the problems that arise from the conflicts between morality and human nature. Policies that try to control human behaviour would be worse because it'd invalidate the very foundation of morality: free will. I'm not sure it'd be worth living a life that has no free will. Worse yet, I challenge you to consider the implications of destroying our free will. Try to figure out how morality can exist without free will. The concept of a moral value we ascribe to rules and behaviours is nonexistent to a life without free will because moral questions must be genuinely deliberated by the consciousness, otherwise the being without consciousness or free will cannot have reasons to judge something as moral or immoral. So taking us in that direction is surely a bad thing. Morality is not about being nice. Morality is about constraining your own evil and creating good through the maintaining of your own vigilance.

[@deus_ex_machina_]

@@cedleonard123 “Whatever ‘hate and intolerance’ mean, they're better than violence” Not always. Violence inflicts wounds of the flesh, which, provided they aren't too severe, heal over time. “Hate and intolerance” often inflicts wounds of the soul, which may not heal so readily.

[@lala45531]

@The Royal Institution , There is a blatant Mitsubishi logo in the middle of your backdrop. Please fix.

[@peterdavidowicz4374]

Requiring a diverse engineering workforce seems impractical for any but the largest companies. One possible solution is government mandated QA questionnaires to a certain proportion of users, with certain diversity response requirements. The fundamental problem is that tech companies are engineering specifically for bottom line advertising and maximum user retention. There needs to be government pressure to force them to also implement user feedback, choice, and control into their practices even if these cut into those other two revenue priorities. Monopoly busting may also increase competition in the long term, but that seems too little too late for the short term corporate tech life-cycle.

[@peterdavidowicz4374]

And also some QA mechanism to force a certain amount of content diversity to be shown, to combat echo chambers and homogenization.

[@menachemsalomon]

There's also the problem in defining "diverse". To some, it means race, gender, or socioeconomic background. To others, it means level of education, socioeconomic status, and political opinions and preferences. In other words, is it diverse if they look different but think alike?

[@crazydragy4233]

​@@menachemsalomon There's also the issue of culture and language, the hardest part to implement imo. This goes back to Tom's point about how things are being fixed by these big platforms in English first

[@nanderv]

A major problem with saying: diverse hiring can solve biases is that you cannot diversely hire people with moral problems with working on AI to monitor platforms. So, you have a bias at KZbin for people who don't mind working for KZbin. It sounds trivial, but this means that the people who choose to work on certain technology are often not the people who hate it or are hurt by it. Even if you have a more racially and gender diverse group of programmers working on the KZbin algorithm, you're still adding people from the bubble of people who love working on AI that can judge whether video content is good or bad, and don't mind that their work is largely judged by whether the advertisers feel trouble or not. I'm not saying: it's bad to have diversity, I'm saying: the diversity you want most is the diversity that is hard to obtain, namely also having people who do not believe in the product be on the team.

[@Littleprinceleon]

Does any mentally sane person with IQ above 80 and age above 12 feel an obligation to follow blindly the YT algorithm? If an adult doesn't know after some years of trial and error what "content" s/he wishes to see and how to find it, the problem is elsewhere... Of course we have limited capacities and most of us will not have such knowledge on any of the topics involving the whole humanity as to be a valuable contributor to the solutions... But that doesn't mean that some educated people can not work on algorithms at least trying to prevent the masses to became a more active part of the problem. Naturally, if the most influential efforts are only (mostly) profit oriented at the cost of more long-term strategies, we can't wonder at the resulting situations

[@mementomori5580]

Why is the Audio so low? I can't understand anything and I already have my audiosettings at uncomfortable levels (any other video will give me hearing damage with those settings and here it's still a quite whisper!)

[@SBImNotWritingMyNameHere]

Bad headphones?

[@davidkrichardson]

Like free speech, it can't be truly free unless you're willing to tolerate some horrific hate speech, but once you start restricting, who (or what algorithm) sets the limits? However, you don't need an algorithm to ferret out truthfulness or set limits. These days, everything is rated, reviewed, certified, ranked, etc. Google could increase rankings for (or allow users to filter) sites based a truthiness rating (rating paid by the content provider - similar to organic food certifications). Yes, there is a tremendous amount of content, but realistically, only a same fraction is seen. I think we can definitely find a way to determine truthiness or at least tag blatantly misleading sites.

[@ajs41]

On the problem of having to use your real name on the internet - what about saying people have to give their real names to the people running a forum or website, but their real names don't actually have to appear next to their comments or posts? Then they would be anonymous to the public, but you would know they're a real person because they had to give their real name to the website or forum in order to be able to post any comments. Obviously you'd have to trust the website/company to keep their details secret.

[@CreativeContention]

It appears to me, that the need for critical thinking in society, is desperately needed more now, than ever before.

[@sybily_du7456]

9:42 This is known as a "Sybil Attack" and is what bitcoin solves by requiring actors to submit a proof of work (somewhat analogous to a captcha exept its for computers not humans) .

[@Rikorage]

Echo-chambers and radicalism are just two sides of the same coin, as was stated in the presentation, which is essentially our tribalism coming to the forefront. If we want to mitigate the damage those kinds of groups do, or to prevent them in the first place, the education of the people to become self-aware of those instincts needs to be an actual focus, and not something you gain knowledge in because of your personal experiences or motivation to be a better person. If the knowledge of who we are is front-and-center, then there should be no need for such extreme tribalism, but I sincerely doubt we'll ever be rid of it, because it is so ingrained within us as a species.

[@rudmanpaul2812]

He's a good man ain't he.... Still think he should get a radio four spot for citation needed

[@DeclanMBrennan]

I kind of like that there is no perfect answer to this just like there is no perfect political system. If humanity ever hits a perfect equilibrium, that would be the end of our great adventure.. Some level of conflict and tension between opposites may well be the engine of human progress and the very essence of life itself.

[@mito._]

I thoroughly disagree. From a scientific perspective. Biologically, we are wired and built to contend with adversarial factors in nature, throughout our lives. We play, we rough and tumble, and we fight. All with the single goal of survival. Survival does not, of itself, endorse progress. Progress is a uniquely human undertaking, which strives to improve the quality of life for as many as possible - for those with access to a better quality of life, typically offered by their civilization. There is no requisite of the necessitation for contempt and competition. With a clear scientific roadmap, the way forward is merely a matter of time. Engineering seeks to find and employ the most optimal solutions possible. Scientists and mathematicians seek to find what is possible, and create the tools necessary to understand it. It then becomes an effort of trial and error, in the truest sense, without the need for trivial human ethics in its way. A system perpetually at war with itself, is no system at all.

[@deus_ex_machina_]

@@mito._ But the rat utopia experiment seems to suggest that without a sense of purpose, society collapses. P.S. I'll probably revisit this comment in a couple years and have no idea what I meant...

[@puig031]

Love the shoutout to tiktok years before it became huge

[@Whyiseverythingthesame]

Hey KZbin, this was Tom Scott talking about tick tock before TikTok was tick tock.

[@manuxx3543]

Maybe platforms should just give the choice to the people Like turning on dark or white theme You could turn on "balance mode, free for all mode, restricted mode" for like twitter or reddit And biased and unbiased when it exist This way people have an imput and can try to master the algorythm, rather than not knowing at all they are nudge into some content or ideas, and it don't drive them out of a platform, just out of a mode of this platform

[@rogueanuerz]

that can create huge inappropriate group if set to restricted mode. and with huge community that would be huge disaster , you know like isis 🤷‍♂correct me if wrong`

[@manuxx3543]

@@rogueanuerz yes it's going into the anonymity problem If it's public it is not really a big deal i think, the law is still there and intelligence watching for dangerous ideas But whether you allow or ban these echo chamber, they will still find a way to form somewhere Maybe it's better to let these echo chamber form where they have an easly acces to other source of influence by looking at another community or switching mode ? I think it will depend on the rules and moderation of each modes, and if different platform would adopt this modele but gives différent rules and moderation, you could see what work and what doesn't But maybe i'm to hopeful to see a big social network adopt that, maybe a niche new one tried already or will do

[@williambarnes5023]

Select censorship level: (◯══) Woke - I am not a responsible adult and am triggered by dissenting opinions. (══ ⬤) Based - I am a responsible adult and can think for myself and make my own decisions.

[@crazydragy4233]

​@@williambarnes5023 The bias in your comment is hilarious

[@williambarnes5023]

@@crazydragy4233 It's also accurate. My goal isn't to be unbiased. It's to be very strongly biased towards reality.

[@liliwheeler2204]

I cannot tell you how confused I was at 4:14 as to why on earth tik tok would have a specific set of guidelines for turkey (🦃)

[@steelbrotherhoodof2359]

truth perceptions are sincerly not universal ? but mostly a unique experience. seldom shareable.

[@steelbrotherhoodof2359]

that is noy science but human. callnit religiouse. ... to end a few discussions. one could say "lets both agree, that we are dissagreeing" "i value your opnion although it is different then mine" or say "lets say we did, but then, we didn't" depending on the listener and or conversatiinal partner. you are doing excellent.

[@57thorns]

15:05 Sir Tom? That is quite the honorific there, or did I miss that Tom Scott had an appointment with Her Majesty Queen Elizabeth II ?

[@rubicon24]

So, Tom... But now that you've mentioned it, it does sound like it.

[@57thorns]

@@rubicon24 *lol* of course. I hope I was entertaining.

[@heidizetzsche8930]

haha, I had the same thought.

[@bob_._.]

Ah, but Her Majesty is not the only person in the world, or even in Great Britain for that matter, authorized to grant knighthood.

[@davidweihe6052]

@@bob_._. This is no longer the 17th, 18th, or 19th century, when anyone could blithely accept nobility titles from anyone (except in the USA), so there was a Prince William of Berlin in the British Royal Family, or Horatio Nelson, Duke of something Sicilian while being a serving Admiral of the British Navy. I don't know the rules for a British citizen accepting foreign titles (and knighthood is technically a title for life, like all future British titles), but it might not be easy. Unless it is a knighthood in the Society for Creative Anachronism or some other reenactment group, of course.

[@Talaxianer]

1:13 is this Daniel Radcliffe in the background?

[@peacewalker3344]

so glad i'm not the only one wondering about that

[@jamesl8640]

Were you not allowed to mention Matt Parker for the stand up mathematician?

[@KufLMAO]

Is this like TED for Europe?

[@u-tubeeditor6696]

[echo - or no boundary]...no boun. is a problem from the point of which it functions from a side but younger/older people find it, from there, what does that mean?...echo is a problem because it will confuse the [enemy] incredibly hard and fairly, only for a moment, then promptly go in the other direction

[@philliphsieh83]

Rii top

[@calmfulspider]

I think requiring specific demographics for a job could lead to bad things. I'm not sure how so, but it's something that worries me.

[@lordchipmonk1]

For starters it opens itself up to criticism of bigotry.

[@hermdude]

14:00 The human element, you say? Better let Todd Rogers hear about that then.

[@primemagi]

Nice to have a bit of honesty & personal fact finding from speaker. the scientist of space & matter tell the tail of current belief(model) and present it as fact. just like priests tailor their talk to matches scripture without question. Ferydoon Shirazi. MG1

[@jaewok5G]

it was dan rather on cbs who was fooled

[@scottmusgrave5735]

Hey Tom, I have a thought on how to root out 'deep fakes'. If there could be some legislation that forces creators; of not only deep fakes, but ALL uploaded content to show the source, that would in effect pull back the blinders. Now, of course, the problem of 'retweets' or sharing content among the masses, could (or would most likely) neglect to offer that legislative failsafe. But it's a start.

[@iluvtacos1231]

How in the world do you enforce that though.

[@petersmafield8722]

As I understand quantum computing will vastly improve the quality of the results from any of the algorithms we have developed for almost all of our search engines. Quantum computing often involves self-learning it would seem to me if we included the instructions to search for actual known facts and having an advanced notice attachment for each post of the quality or quantity of factual information provided in that post. That would allow for fantasy, personal opinion, political opinion, religious opinion, making it clear that the opinions, whatever they are, are in fact opinions, not fact.

[@castiglione4298]

Two minute papers???

[@timq6224]

don't need an algorithm for truth, just need an algorithm to stop the spread of falsehoods. Truth is a combination of fact and perspective. A falsehood ignores one or more contradicting facts.

[@timq6224]

for example. Trump makes the claim that he has created more jobs than any other president (about 7million new jobs before covid) but we quickly find that Obama created 8.5million over the first 3 years of his second term. Therefor, Trump's claim is easily proven as a lie and should be banned. It isn't "opinion" it is an outright lie.

[@ishoottheyscore8970]

@@timq6224 You say that, but beneath a repugnant con-merchants video, I found this an amazing quote to the extent of "When I go to share this video on Facebook, it's flagged as misleading. What is wrong?" even if you algorithmically debunk demonstrable falsehoods, you can't get through confirmation bias and instead feed their paranoia about deep state conspiracies - to them, it's the illuminati, New World Order or Lizard People (or some other euphemism they use to hide their real targets) stopping people from learning the truth

[@steelbrotherhoodof2359]

what makes us suchs a wiseguys... we are older then you... it is not our exquisited mambo jambo. it is just life experience. (and getting a lot of bleu ones (bleu notes ?) in bars, streets, forests, ally's schools, shops, musea..... )

[@steelbrotherhoodof2359]

nobody likes a wiseguy...

[@pebblecups]

Sitting on a sacred relic.

[@DeclanMBrennan]

That table top has no doubt been polished by many venerable posteriors from Faraday onwards.

[@thatguyoverthere531]

What we didn't see is Tom asking if it was ok for him to sit on it.

[@Markle2k]

@@DeclanMBrennan According to the RI's response to another relic worshipper, it is also the third replica of the original desk. "A desk is made to be used! This is the third copy of the original Faraday desk and each of them has seen chemical splashes, explosions and many famous bums."

[@edonslow1456]

Watching this 3 years on its interesting. AI has really taken off since then, and AI content moderation is almost inevitable.

[@neneppie]

We might've reached that point where deepfakes are indeed a significant threat. Only a matter of time now before we begin to *see* the consequences

[@jamesl8640]

I love how Tom Scotts most viewed video is sending a garlic bread to up

[@philliphsieh83]

Thus saith the Lord!

[@celinak5062]

10:40

[@rhamph]

At a fundamental way we need an intelligence that can make reasoned decisions on what is Good and what is Bad for humanity. Today that intelligence is us and although deeply flawed we seem to have accumulated incremental improvements over the centuries. In that future that intelligence may be artificial but I can't help but wonder what improvements we could make to ourselves if we tried.

[@menachemsalomon]

The problem with that is that your definition of Good and Bad may not comport with my definition. That delineation must precede the intelligence that can divide things into Good and Bad based on any particular definition.

[@artski09]

Good and what is Bad for humanity sounds a bit grand or quite big stakes for a chat

[@musikSkool]

(Not that he will read the comments) What do you think would happen to the world if all advertisements were banned?

[@isaacorellana1754]

Regression.

[@menachemsalomon]

The company might fold or simply close. Keeping the platform viable has costs that must be met somehow.

[@Skooteh]

confusion. Advertising exists to inform the customer. Having a nice label is advertising. are labels now banned? it would mean there's a lot more effort necessary to find any given product you're after. Also, so many industries that rely on advertising will just die.

[@musikSkool]

I'm pretty sure most websites would switch to subscription-based services. Or you would have an online "wallet" and every page you visit, your browser would automatically give them a penny. Your internet service provider would handle all that for you. 1,000 webpages = $10

[@musikSkool]

@@Skooteh No, labels are not advertisements, they are logos on products. Advertisements are billboards or video commercials. A company can still put a logo on a product for sale, or their building.

[@evanplanas]

so the answer is quota systems and affirmative action.....good luck.

[@Flocksta]

9:30 that guys is not happy with that question :D

[@QuixoteX]

Sponsored by Mitsubishi?

[@tkejlboom]

Is there a term for the frisson of cognitive dissonance when someone calls for a more diverse and cosmopolitan constituency than tech and then suggests Washington DC? I think the elephant in the room is what happens when the gatekeepers are broadly reviled and mistrusted.

[@FinetalPies]

Having a Q&A is kind of interesting. Obviously Tom is well researched, but he's still not an expert?

[@FinetalPies]

I definitely share the feeling that his opinion is one I value, but that feels like more and more of a dangerous proposition these days.

[@christianbarnay2499]

You don't need to be an expert to give some insightful clues to a subject. Especially when you're honest enough to say "I don't know" or "there are other persons better qualified than me to answer that subject". He's giving us starting material to bootstrap our own thinking and research on that complex subject.

[@MrGundawindy]

You think the big tech companies have a moral imperative to not let the Nazis in but also are concerned with echo chambers lagging to radicalisation. That seems somewhat contradictory. Wouldn't it be better to allow everyone to speak freely so bad ideas can be countered with better ideas rather than creating these echo chambers by restricting certain groups free speech?

[@Anon0nline]

Sir Karl Popper wrote the equation for Rational Conclusion. Tom Scott is not an epistemologist, scientist, or mathematician.

[@sweetspotendurance]

14:02 Todd Rodgers has entered the chat

[@kaitlyn__L]

I recognise the voice of the guy at 5:10 from another Q&A.. but I seem to recall he was, at that time, arguing that bias isn't real. Interesting reversal of position.

[@RedShirtGuy96]

The ability to argue for or against one position effectively regardless of your own personal beliefs is skill common in academic settings. It is a very useful skill for critical thinking. Just my thoughts on why this might be the case here.

[@shexdensmore]

What are you talking about when you speak of Echo chambers

[@ammonkunzler3948]

Communities where everyone agrees on something and dissenting voices are prohibited, so the same ideas are bouncing around forever like a like a literal echo chamber.

[@bknesheim]

I just wonder why the add before and at the end was for "New Barbie Dolls".

[@PastPresented]

Ain't algorithmic targeting wonderful 🥴

[@bknesheim]

I was around when Barbie was a new thing, but was not then nor now, in the marked for one. It really make me wonder if the algorithm is much better then just random.

[@MLIOGJXNUYAT]

No, "the INTERNET interprets censorship as damage and routes around it". The WEB is something else entirely.

[@ThatisnotHair]

Chads know there is no truth

[@andybaldman]

*Ditch recommendation systems. Return to organic search, and force people to find their own stuff. Has anyone else noticed how nobody uses search anymore? Facebook is full of people asking the same inane questions over and over, without having ever tried a 10 second google search. The old internet was not like this, and people got spanked for not 'searching before posting'. What changed?*

[@christianbarnay2499]

Recommendation systems when used appropriately are a very powerful way of getting tedious tasks done fast and saving plenty of time for more interesting activities. The problem you're describing has nothing to do with the tool, the real issue is people being lazy and waiting for others to feed them with answer without even having to ask the questions. The only solution is to teach people their brain is not a passive memory sponge. And that a single isolated opinion, even when expressed by their best friend, should never be considered a universal absolute truth.

[@davidweihe6052]

Asking repeated questions on Facebook is no different than asking them from Google, and then using that to drive the "organic" search, except that you use Google's algorithm's inherent biases rather than, frex, KZbin's.

[@andybaldman]

@@davidweihe6052 In the Google case, people are helping themselves (vs FB, where they're asking others to spoon feed them, from knowledge other people already have). However in the FB case, the forums get clogged with the same questions over and over, making it inefficient for all other readers.

[@andybaldman]

@@christianbarnay2499 The problem is, the reco systems have made people lazy. And the entire internet has gotten dumber as a result. The systems are only powerful when there are humans in the loop. However reco systems make the humans become passive (vs an active participant in the search). So the system loses intelligence, and the people become passive and controlled (i.e., they'll watch whatever is recommended, vs actively seeking out new things.)

[@christianbarnay2499]

@@andybaldman The reco system doesn't make people lazy. Lazy people have made their own decision to just follow the reco and stop looking for other options. As long as the search bar exists they have the option to seek various information sources by themselves. Lazy people are responsible for their own laziness. Blaming the reco system is just an excuse.

[@AtheistEve]

I’m not subbed to Tom or RI. How did I get here? I’m not interested in homeopathy either.

[@JoaoPedro428]

Maybe you watch a lot of tech videos?

[@AtheistEve]

Joao Pedro No, I don’t watch that many tech videos. I’ve turned off the recommended videos option. So I’m starting to see a wider variety coming through on my home page.

[@bob_._.]

@@AtheistEve One would assume you got here by clicking on the link to this, one of the wider variety of videos that have been showing up on your home page. Unless you have autoplay turned on and let YT do the clicking for you.