How to train your social media algorithm puppy to fetch whatever you want online - and request to social media companies to give us ability to select between - pretrained options

I’m a fact checker for a group on Facebook that helps people who get scared of many things shared on the internet.

Nowadays a lot of the people we help ask for us to fact check things that appear in their social feed. They tell us they DON’T WANT TO SEE THESE THINGS but they can’t stop them and they keep scaring them.

The only solution is to give up on social media which they don’t want to do especially if it is one of the main ways they interact with people online. For some people it is their main way of socializing nowadays.

This post will

• help you to configure your own social feed to be more what you want

• is a call out to social media companies to give us options to use check boxes to decide what we want to see

• which will lead to far happier customers and eliminate almost all the mental health issues arising from social media feeds

This shows how the check boxes could work

TEXT ON GRAPHIC: Mockup - proposal for social media companies

On the left, "Social feed" panel with three stacked posts, each featuring a photograph of a wind surfer riding waves on a sunny day.

Each post includes icons for likes, comments, and shares.

To the right, a "Custom" filter panel with a vertical list of checkboxes for different content categories.

Only "Please show me photographs of colourful birds and check to make sure they are not AI generated" and "Wind surfing" are checked;

all other options, such as "News and analysis," "Personal news from family and friends," "Astronomy," "Flora and fauna," "Conservation," "Anime," "Computer games," "Table tennis," "Gardening," and "Music (genre)" are unchecked. The list extends out of view, suggesting more options are available.

Blue text to the right and below explains:

Text next to the check boxes editable

• used as instructions for chatbot to curate feed

• can check or uncheck as many as you like

• In this way can configure your feed as you like when you like

Option to switch between lists like this for:

• research

• news

• recreation etc

Idea for a way to make social media far better for mental health, while also increasing engagement of the right sort for the companies hosting them.

• fun and much more productive and enjoyable to use.

Graphic made with Bing copilot https://copilot.microsoft.com/shares/cEPKKhGU7B5snZ8EHDP1X

Meanwhile here is how to get the same effect yourself - but with much more work, and only able to configure it one way at a time.

Panicking people engage very strongly with scary things.

If you click on one scary post out of hundreds it will go fetch more and more of them.

Be consistent. NEVER click on anything scary.

Your social media algorithm is like a marvelous puppy - you can train it to fetch anything from billions of posts on Facebook.

But it doesn’t know what you want to see.

It can only go by your clicks, likes, the reels you play etc.

To train it, click, like, love, play WHAT YOU WANT TO SEE

Photo: Puppy wuppy, a photo of puppy called Roxy by Michael.

I trained my algorithm puppy on Facebook to fetch photos of colourful named wild birds and beautiful nature scenes. Named wild birds because the unnamed ones nowadays are often fake AI and I don't want to see fake AI willd birds.

It only took a few days of consistently not responding to anything else.

Click on them. Like. And if especially keen on it then love react.

I now have added a few other things e.g. corrugated iron buildings.

And then VERY CAREFUL not to click on anything you don't want to see.

TEXT ON GRAPHIC:

Facebook feed - a place for fun stuff.

NOT FOR NEWS. If it looks like news - to say “no” don’t click, read, play, like, nothing.

This is a screenshot from part of my own feed.

the two items are: https://www.facebook.com/groups/naturewildlifebirdphotography/posts/1619754655381283/

and https://www.facebook.com/groups/1233601416666031/posts/29008557215410411/

Notice that I liked them both. That is how you tell the algorithm what to get for you.

The algorithms today are marvellous and very effective.

How I trained Twitter to fetch reliable news reporting on topics that interest me from experts- you can train the algorithm to fetch reliable news but hard to do and not always consistent

So what about news? It’s usually just called a “feed” now. But often referred to as a “news feed”.

But despite the name, social media is not the place to find out good information about things that scare you, e..g warss.

This only works if you very carefully train your algorithm dragon to find only reliable news which is not easy to do.

For easily scared people it is best to train it not to find news at all and instead get your news from more reliable sources.

However you CAN train your algorithm dragon to fetch reliable news, at least on Twitter.

I use Twitter differently from Facebook. There I follow experts on climate change, infectious diseases, the Ukraine war, Gaza strip, the US election now, and so on. If an expert shares something by another expert I check out their twitter feed and follow them too if they are suitable. And I click almost only on the things I want to see there - though I also click on a few fun things for relaxation and to break it up, but not too many as I don't want my feed to fill with fun stuff.

In this way I have a Twitter feed that is superb for finding the latest on global warming and action on climate change, the election, on the Jan 06 / presidential immunity legal proceedings, on the Ukraine war and Zelensky's victory plan, the latest in Israel / Lebanon, I still get updates on COVID and on masking etc etc. Just by going to my news feed on Twitter from time to time I don't even need to search for it.

But this requires you to be very disciplined. It becomes second nature once you do it. But it is so easy to get distracted even knowing what this will do to your news feed. Luckily Twitter responds very quickly and if you start to get lots of stuff you don't want to see, then just not responding to any of it and it quickly disappears again.

It is also never 100%. I get a few mistakes shared in my feed. There are some people who are very responsive to the latest developments in the Ukraine war for instance but they don’t check things and sometimes share unconfirmed rumours that turn out to be false. Other times the same people may be the first to alert me to a major new development. So when I see something from one of them I search for confirmation from someone more reliable.

So - it isn't necessarily about avoiding all news on social media. It can be very good.

I find Twitter especially useful for the Ukraine war, if I only had access to mainstream news I'd be very ill-informed about it. Almost all that I know about it now comes from experts on Twitter.

If I only had mainstream news I'd have no idea how weak the Russian position is in Ukraine. That is based on BOTH Ukrainian AND Russian sources including many people sharing translations of whatever ordinary Russian soldiers,.officers and milbloggers are currently talking about on Telegram.

I'd have no way of knowing this without X / Twitter and then to some extent Institute for Study of War - and other sources.

This is an example to show how valuable social media can be for fact checking and news too not just recreation but you have to get it right.

Most people will want to train their algorithm to not fetch any news ever

So for most people social media is NOT a good place to get news. Sadly you need to TRAIN your social media dragon to fetch you what you want to see and not ever fetch you news.

Social media companies are losing customers by trying to boost engagement - need ways for their customers to be safe and happy

Focusing on engagement misses the difference of the type of engagement. Is someone panicking going to be clicking on ads? Plus longer term it leads to many just abandoning social media altogether. For long term engagement they need social media to be a place where people feel safe and happy.

I think Social media need to let us configure our social feeds. If we want to see news, great, have a check box to include news. But even those who use e.g. X to follow fast developing stories and analysis by experts as I do, there are times I might want to go to X and e.g. find out about astronomy or just to look for relaxing images e.g. of flora and fauna in my case or nature. So - we should be allowed to configure what we want to see at any time.

Proposal for social media companies - select what you want to see in your social feed categorized by interests

This is the graphic that we saw at the beginning of this blog post - let’s look at the mockup more closely.

TEXT ON GRAPHIC: Mockup - proposal for social media companies

On the left, "Social feed" panel with three stacked posts, each featuring a photograph of a wind surfer riding waves on a sunny day.

Each post includes icons for likes, comments, and shares.

To the right, a "Custom" filter panel with a vertical list of checkboxes for different content categories.

Only "Please show me photographs of colourful birds and check to make sure they are not AI generated" and "Wind surfing" are checked;

all other options, such as "News and analysis," "Personal news from family and friends," "Astronomy," "Flora and fauna," "Conservation," "Anime," "Computer games," "Table tennis," "Gardening," and "Music (genre)" are unchecked. The list extends out of view, suggesting more options are available.

Blue text to the right and below explains:

Text next to the check boxes editable

• used as instructions for chatbot to curate feed

• can check or uncheck as many as you like

• In this way can configure your feed as you like when you like

Option to switch between lists like this for:

• research

• news

• recreation etc

Idea for a way to make social media far better for mental health, while also increasing engagement of the right sort for the companies hosting them.

• fun and much more productive and enjoyable to use.

Graphic made with Bing copilot https://copilot.microsoft.com/shares/cEPKKhGU7B5snZ8EHDP1X

The first two would be for everyone:

News and analysis []

Personal news from family and friends []

Both of those would be configurable - sources to use for the news, topics that interest you - which of your Facebook friends you want to hear from etc

That part is not show in this simple mockup.

The other options could be auto generated from interests shown by your activity on the social media platform.

E.g:

Astronomy []

Flora and fauna []

Conservation []

Anime []

Computer games []

...

A table tennis fan might have

Table tennis[]

A windsurfer might have

Wind surfing[]

Others might have

Gardening []

Music (genre) []

Also, nowadays with chatbots you could write a long form description of what you want

"Please show me photographs of colourful birds and check to make sure they are not AI generated".

The simplest way to do it would be to just make all the text fields next to the check boxes editable, so you can type in anything you like.

In graphic I have a mock up of what it looks like after the user edited the third check box to read:

"Please show me photographs of colourful birds and check to make sure they are not AI generated"

This is well within the capability of social media algorithms.

As explained in this post, I was able to train Facebook to show me just photographs of:

• Colourful birds [not AI]

• insects, flowers, fauna and flora, sea life, relaxing scenes from nature.

I later added

• corrugated iron sheds

• Recorder music

Just by constantly clicking on only those things and only liking and reacting those things.

But that is very limited because maybe some day I want to see something different and I can only do that by spending a couple of days retraining the algorithm

So - I think configurable feeds would work FAR BETTER in terms of

• happy customers

• long term engagement

• new customers as the reputation of the social media spreads for being a nice place to be

• not losing customers.

Also, people in a relaxed frame of mind knowing what to expect in their feed are far more likely to click on ads too.

If you are interested in gardening and want to buy seeds or plants or gift plants for a friend - are you more likely to do it

• if you can configure your feed to show plants, flowers, gift plants, seeds etc

• if the feed is full of reels claiming a world war or human extinction?

I think it is obvious which one would work best.

Right now the way the algorithms work - though sophisticated in some ways - in other ways is so primitive

They have these metrics where they just measure how many hours you spend online and how often you click on / like / engage with posts and don't take the wider or longer term view.

They decide what to show you with no input from you.

I think future users of social media and social media companies will find this approach bizarre. Why were we not given any control of our feeds?

So I think this would help social media too and the first social media company to do this would likely get an advantage over others that don't.

Or it could be required by law.

What you can do in the meantime - how to train your social media puppy / dragon

TEXT ON GRAPHIC: HOW TO TRAIN YOUR (ALGORITHM) DRAGON

The important thing is to be consistent when training your algorithm dragon

If you can’t right away try to do this for a few hours, then a day, then several days.

Your social media algorithm dragon can be your friend.

1. CLICK on what you want to see

2. LIKE, or LOVE react to boost it

If you don’t want to see something:

1. STOP and ignore

2. NEVER click on it or look at comments or share or engage

. Hiccup and Toothless

Details see:

CONTACT ME VIA PM OR ON FACEBOOK OR EMAIL

You can Direct Message me on Substack - but I check this rarely. Or better, email me at support@robertinventor.com

Or best of all Direct Message me on Facebook if you are okay joining Facebook. My Facebook profile is here:. Robert Walker I usually get Facebook messages much faster than on the other platforms as I spend most of my day there.

FOR MORE HELP

To find a debunk see: List of articles in my Debunking Doomsday blog to date See also my Short debunks

Scared and want a story debunked? Post to our Facebook group. Please look over the group rules before posting or commenting as they help the group to run smoothly

• Facebook group Doomsday Debunked

Also do join our facebook group if you can help with fact checking or to help scared people who are panicking.

SEARCH LIST OF DEBUNKS

You can search by title and there’s also an option to search the content of the blog using a google search.

CLICK HERE TO SEARCH: List of articles in my Debunking Doomsday blog to date

NEW SHORT DEBUNKS

I do many more fact checks and debunks on our facebook group than I could ever write up as blog posts. They are shorter and less polished but there is a good chance you may find a short debunk for some recent concern.

I often write them up as “short debunks”

See Latest short debunks for new short debunks

I also tweet the debunks and short debunks to my Blue Sky page here:

. doomsdaydebunks.bsky.social

I do the short debunks more often but they are less polished - they are copies of my longer replies to scared people in the Facebook group.

I go through phases when I do lots of short debunks. Recently I’ve taken to converting comments in the group into posts in the group that resemble short debunks and most of those haven’t yet been copied over to the wiki.

TIPS FOR DEALING WITH DOOMSDAY FEARS

If suicidal or helping someone suicidal see my:

BLOG: Supporting someone who is suicidal

If you have got scared by any of this, health professionals can help. Many of those affected do get help and find it makes a big difference.

They can’t do fact checking, don’t expect that of them. But they can do a huge amount to help with the panic, anxiety, maladaptive responses to fear and so on.

Also do remember that therapy is not like physical medicine. The only way a therapist can diagnose or indeed treat you is by talking to you and listening to you. If this dialogue isn’t working for whatever reason do remember you can always ask to change to another therapist and it doesn’t reflect badly on your current therapist to do this.

Also check out my Seven tips for dealing with doomsday fears based on things that help those scared, including a section about ways that health professionals can help you.

I know that sadly many of the people we help can’t access therapy for one reason or another - usually long waiting lists or the costs.

There is much you can do to help yourself. As well as those seven tips, see my:

BLOG: Doing what matters at times of stress
— new WHO booklet to help with mental health during COVID19 pandemic

BLOG: Breathe in and out slowly and deeply and other ways to calm a panic attack

BLOG: Tips from CBT
— might help some of you to deal with doomsday anxieties

PLEASE DON’T COMMENT HERE WITH POTENTIALLY SCARY QUESTIONS ABOUT OTHER TOPICS - INSTEAD COMMENT ON POST SET UP FOR IT

PLEASE DON'T COMMENT ON THIS POST WITH POTENTIALLY SCARY QUESTIONS ABOUT ANY OTHER TOPIC:

INSTEAD PLEASE COMMENT HERE:

Post to comment on, with off topic potentially scary comments - or send me a private message - or use our group on Facebook

The reason is I often can’t respond to comments for some time. The unanswered comment can scare people who come to this post for help on something else

Also even an answered comment may scare them because they see the comment before my reply.

It works much better to put comments on other topics on a special post for them.

It is absolutely fine to digress and go off topic in conversations here.

This is specifically about anything that might scare people on a different topic.

PLEASE DON’T TELL A SCARED PERSON THAT THE THING THEY ARE SCARED OF IS TRUE WITHOUT A VERY RELIABLE SOURCE OR IF YOU ARE A VERY RELIABLE SOURCE YOURSELF - AND RESPOND WITH CARE

This is not like a typical post on substack. It is specifically to help people who are very scared with voluntary fact checking. Please no politically motivated exaggerations here. And please be careful, be aware of the context.

We have a rule in the Facebook group and it is the same here.

If you are scared and need help it is absolutely fine to comment about anything to do with the topic of the post that scares you.

But if you are not scared or don’t want help with my voluntary fact checking please don’t comment with any scary material.

If you respond to scared people here please be careful with your sources. Don’t tell them that something they are scared of is true without excellent reliable sources, or if you are a reliable source yourself.

It also matters a lot exactly HOW you respond. E.g. if someone is in an area with a potential for earthquakes there’s a big difference between a reply that talks about the largest earthquake that’s possible there even when based on reliable sources, and says nothing about how to protect themselves and the same reply with a summary and link to measures to take to protect yourself in an earthquake.