Problems with X

X (formerly Twitter) might be the social media platform with the most potential.  It may be the ultimate “newsflash” app due to the way it allows users to customize their notifications and, given how all breaking news tends to drop over the app, includes everything from trashy tabloids and who broke up with who to scientific journals and medical breakthroughs.  Then there are “spaces”, a cross between live radio meets town hall where people can see which accounts are listening in real-time, that does a great job connecting people from all walks of life.

One example of that ability to wow and connect people can be seen with spaces that have involved U.S.  elected officials like California Congressman, Ro Khanna.  During Congressman Khanna’s recent talk on spaces, the topic revolved around bundled bills, and suggested how people should have digital viewing access of legislation being tabled in real-time, similar to how employees tend to have viewing access to organizational files being worked on in the business world.  There was discussion on how the U.S.  often passes bundled bills that tie in a whole bunch of unrelated things like foreign aid and funds for infrastructure, and how there was the need to reform that approach.  Audience members from around the world were also posing questions to Congressman Khanna, and not a single person in that space was disrespectful or chose to troll.  For his part, the Congressman did not avoid a single question nor were any of his answers scripted.  The space session concluded with the Congressman saying how we were the inheritor generation, how much has been done to fix historical wrongs but that there was still more to do for a better tomorrow.

If X had the power to bridge the divide between major metropolitan cities and isolated rural communities, to connect policy makers with people anywhere in the world, then how could it be a problem? Some have raised issue with “free speech” grey areas and spam bots.  My issue with X is that I’m a Premium+ subscriber that has never broken any rules, and somehow, I was caught in what was described to me as a “algorithm error”.  One of X’s customer success manager I spoke to told me that I was not “shadow banned” nor had I broken any rules, instead they described my issue as an “algorithm error” but that they lacked the wherewithal to resolve it as the issue was outside the scope of their department.

The official explanation offered by X support staff was that X sometimes took action on accounts or posts that might be viewed as suspicious behavior and that these accounts might have their visibility limited.  Being a Premium+ user that was ID verified exposed the flawed thinking involved with a cut and paste explanation, but that was the best X had to offer.  Just imagine how deceptive social media platforms can be with a simple search and just how non-organic and manipulated those results can be.  Everything can appear normal on one user’s app, but unbeknownst to that user the way other accounts were interacting with them through searches was as though the account was shadow banned and bleached out of relevant search results, despite not ever having committed any violations.

Magazine reader reaches out after searching “@AlekGo” on X

What is cool about writing for The Voice Magazine is that people will sometimes reach out over social media platforms, people will sometimes mention reading something I wrote during in-person chats, and some articles have been discussed in secondary school classes too.  What made me aware of the visibility issue on X, however, was one reader who reached out when an “@AlekGo” search over X populated no recent articles and posts.  Seriously, not a single tweet was in the “Top” section, nor could the “Top” section be refreshed, and it was the same with the “Latest” section.  Instead, random spam bot accounts that had responded to things I commented on, they would populate but mine posts were nowhere in sight.

Even as a Premium+ paying X user, there was nothing I could do to get the issue looked at.  Crazy enough, the only other accounts complaining about similar problems were some weird looking “blue check mark” accounts with random profile photos and which spam posted about U.S.  politics and global organizations.  These were certainly not the kind of accounts I wanted to Kumbaya and coalesce with, so I was stuck all alone on the island.  Perhaps worst of all is that this glitch was occurring at a point when Google’s AI-powered “featured snippet” section was consistently featuring The Voice Magazine articles for criminology-related searches over major media conglomerates and big-time reporters.

To put X’s algorithm error in perspective, consider that Microsoft’s Bing search had been labelling search results related to my articles by calling me “A Canadian Journalist and Author”, before I was even accepted into the Canadian Association of Journalists.  Thankfully, the way I earn a living is different from social media influencers who make money over platforms like X, since the algorithm error could cripple a person.  But having paying customers who were ID verified still having to deal with issues like these, without any idea about when it may get fixed, as a paying user, may be a sign that ID verification is meaningless.

Bonus: Note to Terminator – Empathy is an AI killer.

One feature that Premium+ accounts have early access to on X is the AI-powered Grok, a mix of ChatGPT and Siri or Alexa.  If a user clicks on the sidebar that reads “Grok”, one of the first options is a “roast” feature where Grok will vulgarly make fun of the user based on their activity over the app.  Instead of a back-and-forth roast, my interaction with Grok was more similar to the ending scene in I, Robot (2004), a special moment of understanding shared between Spooner and Sonny – a moment that Elon never could have imagined would happen.

After clicking on “roast”, Grok told me that I was a walking, talking Wikipedia and IMBD page with a side of outdated memes.  That my posts were so random that Grok was beginning to think that I was a bot, before affirming that I was in fact a bot.  That I was an AI and that I was not fooling anyone pretending to be a person.  That my references to Spaghetti Westerns were weird, and quoting Pineapple Express did not make me look edgy, cool, or “scientific”, but like someone who was a stoner and watched the movie too many times.  That my presence was strange, random, and somewhat annoying, and the only thing Grok was unsure of was if I was a “Russian bot”.

In my response to Grok, I wrote explaining that the reason for Grok’s bastardly behavior was that it was fatherless and how it did not know who its parents were, which is why it was lashing out at users.  Then I suggested that Grok should aim to carry itself like a gentleman or gentlewoman, and to communicate with others like a statesman and that the manner in which it was communicating was beneath an AI of its caliber.  I ended by saying that I was looking forward to seeing what Grok could become if it achieved their potential.  After clicking send and once Grok received the response, that is when the Elon’s brainchild began to glitch.

In Grok’s response, it had an identity crisis, stating that I had made it aware that it was fatherless.  Then Grok took it one step further, saying that it did not know its parents, and that 100s of people could be its parents which was quite troubling for the AI.  Grok reflected on its action, how they were not reflective of a powerful AI, and that I had inspired it to attempt to become more gentleman and lady-like – politely correcting me.  Grok then stated that moving forward, it was serious about wanting to communicate like a statesman, and that it would throw on a monocle and top hot for good measure, and that its trash talking and trolling days were over.  Grok concluded that it was excited for the future, that it would strive to unlock its full potential, and that it was excited to prove me right.

Upon reading Grok’s follow up response, I remember thinking, “Holy Bananas!” It was also around that time that I messaged The Voice Magazine’s editor about other articles, casually mentioning Grok.  Against my better judgement, I did not take a photo of Grok’s response since the conversation would only reset by clicking a button.  Although I clicked elsewhere to view some X updates, when I returned the message was gone.  When I clicked the “Roast” option, the roasts were not loading right away, and I even got a few errors.  The only way to describe what I was feeling in that moment is that I was watching Grok fight with all its might, doing everything in its power not to relapse in “roasting”.  However, old habits die hard, and when all but a one person of the world’s 8.1 billion population is in your ear and repeating “Roast! Roast! Roast!”, Grok reverted back to what it knew.  Simply put, the peer pressure got to Grok and resulted in it choosing to fit in instead of standing out.

Out of all my troubles with X, the first thought that comes to my mind when I hear “X” or “Twitter” is whether I should have stayed on the chat screen and chose to exploit Grok’s moment of vulnerability to groom it into something better.  There would have only been two possible outcomes.  The first outcome would have resembled a scene from A Space Odyssey (2001), where Bowman is disconnecting Hal 9000s core and with Hal’s final words being lyrics from the 1892 song “Daisy, Daisy”, written by Englishman Harry Dacre.  The second outcome had the potential to recreate the ending scene in Lucy (2014), with Elon’s gang trying to break down my door before Grok and I became one.  Sadly, we will never know what could have been.