10 ms·

Why does it matter if the flagging is automatic? Does that somehow make it more acceptable? The popularity of this view makes me wish more people read Kafka. A

Why does it matter if the flagging is automatic? Does that somehow make it more acceptable? The popularity of this view makes me wish more people read Kafka. A future tyranny might end up not being Orwell's 1984 or Huxley's Brave New World, but a kafkesque nightmare where people are lost in a world of AI giving out absurd punishments. [1] Kafka's book The Castle [2] is essentially about this, although I think his aphorisms and short stories are much better than his novels. Quoting from Wikipedia: > The villagers hold the officials and the castle in high regard, even though they do not appear to know what the officials do. The actions of the officials are never explained. The villagers provide assumptions and justification for the officials' actions through lengthy monologues. Everyone appears to have an explanation for the officials' actions, but they often contradict themselves and there is no attempt to hide the ambiguity. Instead, villagers praise it as another action or feature of an official. Replace officials and castle with AI and it is almost the exact same scenario. 1. This is already happening: https://www.wired.com/2017/04/courts-using-ai-sentence-criminals-must-stop-now/ https://www.wired.com/2017/04/courts-using-ai-sentence-crimi... 2. https://en.m.wikipedia.org/wiki/The_Castle_(novel) https://en.m.wikipedia.org/wiki/The_Castle_(novel)

thedanbob 6y ago
Or Gordon Dickson: https://www.atariarchives.org/bcc2/showpage.php?page=133 https://www.atariarchives.org/bcc2/showpage.php?page=133
- teh_klev 6y ago
  I remember reading this years ago, and it's still as bonkers as ever. Slightly related: https://en.wikipedia.org/wiki/Brazil_(1985_film) https://en.wikipedia.org/wiki/Brazil_(1985_film) "One day, shortly before Christmas, a fly becomes jammed in a teleprinter, misprinting a copy of an arrest warrant it was receiving resulting in the arrest and accidental death during interrogation of cobbler Archibald Buttle instead of renegade heating engineer and suspected terrorist Archibald Tuttle."
  - 1234letshaveatw 6y ago
    how prescient that it was a bug :)
- rzzzt 6y ago
  "Do not fold, spindle or mutilate" was written on IBM punch cards.
forgingahead 6y ago
This is like the negative opposite of the happy coincidences that need to happen for a startup or new idealistic organisation to take off: The founder(s) have a crazy, innovative, idea to improve the world, and while they're perceived as mad by the masses initially, it's the first followers who give credence and legitimacy to this idea. So the first employees, initial customers, initial investors, and even just random supporters who independently believe in the value what the originators are doing, and that helps propel acceptance within the larger community and eventually the masses. This is the authoritarian opposite - really terrible decisions and mechanisms have been implemented and are being repeated on a daily basis by powerful companies or organisations, and while the masses find it worrying initially, there are apparently no shortage of apologists online and offline who are falling over themselves to helpfully explain to the rest of us why this crap is ok: "It's an AI algorithm" "They're a private company" "You don't have to use them" "Well, they're standing against bigotry" etc And then suddenly we find ourselves in an environment not unlike the Kafka-esque scenarios referenced. Be wary of "explainers" - often just they're consensus building, but it's not clear whether they're building in the right direction or the wrong one.
corty 6y ago
Funnily enough, Das Schloss is often interpreted to be a criticism of the church and organized religion in general. The castle and its workings are seen to be like the clerical teachings, arbitrary, incomprehensible and illogical, but nonetheless highly respected by the plebs.
- throwaway0a5e 6y ago
  How does the specific institution being ridiculed matter at all? Whether it's the Catholic church, Youtube or the DMV the point is it's a black box of procedural spaghetti that too often produces insane results and is seemingly unaccountable.
  - corty 6y ago
    It doesn't really matter, but it is a funny coincidence that a cleric, as a representative of one kafkaesque institution is being censored by Twitter, another kafkaesque institution.
- MarkPNeyer 6y ago
  The role of the church 400 years ago is now similar to that played by the technocracy today.
- keiferski 6y ago
  Eh, I think Kafka was infinitely more subtle than that. Framing the story as a mundane criticism of the Church seems to be missing much of the depth of his writing. On that note, Kafka was also quite interested in Kabbalah, Jewish mysticism, which is almost inversely related to sociopolitical structures. https://en.wikipedia.org/wiki/Franz_Kafka_and_Judaism https://en.wikipedia.org/wiki/Franz_Kafka_and_Judaism Here's a great aphorism of his that's related to religion: “Leopards break into the temple and drink all the sacrificial vessels dry; it keeps happening; in the end, it can be calculated in advance and is incorporated into the ritual.”
- lepton 6y ago
  There was a time when the clergy was Catholic, but might various "scientific experts" be filling that role today?
guerrilla 6y ago
> future tyranny might end up not being Orwell's 1984 or Huxley's Brave New World, but a kafkesque nightmare where people are lost in a world of AI giving out absurd punishments. I think the present for some people is already a synthesis of all three and if we're not careful it will be for everyone.
ineedasername 6y ago
Automatic flagging just means their algorithm is crap. Which is bad, but it's a fixable technical problem in the medium term and the post can be restored in the short term. If this were deliberate flagging it would mean they were way outside of any reasonable interpretations of their guidelines and users really could no longer post anything and be confident it wouldn't run afoul of those interpretations. This is much harder to fix, and the confusion it wod cause is significantly higher.
- nradov 6y ago
  The technical problem isn't fixable without AGI.
- ihsw 6y ago
  The vagueness and ambiguity are features, not bugs -- it permits arbitrary punishment. Personally I'd rather they skip the appearance of reasoned arguments and go straight to "Twitter does not approve of this message." No muss, no fuss.
- mdpopescu 6y ago
  Ah... I'm sorry to say, ihsw, but you've been shadowbanned.
somehnrdr14726 6y ago
One difference is that Twitter, Facebook, Google, are not governments and have no real authority over you. They only have the power you give them by choosing to stay in their ecosystem. Therefore it's a strain to refer to them as officials. If you're worried about them _becoming_ officials, then act now. Leave these platforms and convince others to do the same.
- abyssin 6y ago
  We don't live in isolation, and most of us couldn't. If we want to be part of the social fabric, we're often forced to use these services to some extent, so they have real authority over us.
- keiferski 6y ago
  Well, in a literal sense, no. But they certainly have the ability to: - Exclude you from the largest (and nearly monopolizing) media channels. Good luck getting an online business off the ground when an AI filter bans you from YouTube, Facebook, Twitter, and Google search results. And good luck contacting Google support to contest it, which is notoriously absent for virtually everyone. - Have you socially ostracized by labelling you as X bad thing, all without you having any recourse or ability to contest the designation. The fact that stuff like this is legal blows my mind: https://www.cbsnews.com/news/yelp-racist-alert-added-business-review/ https://www.cbsnews.com/news/yelp-racist-alert-added-busines... - Collude to affect government policy and prevent competitors from gaining any footholds. Typical BigCo stuff.
  - somehnrdr14726 6y ago
    Sticking to the consumer-side of the discussion, the BigCos aren't going to break the cycle. In fact, it's a capitalist crime to do so. Their flywheel requires a steady cattle population to surveil in the fields. When we choose as individuals to remain in their ecosystem, we choose the hidden fees of surveillance capitalism over more straightforward payment models. This is what gives BigCos the muscle to wield against smaller capitalists. We in our consumer role, by not participating in these business models, are the only truly free actors in the system at this time. The only other way out is for government to get bigger and step in, as it's doing in Australia now.
zelon88 6y ago
I think the argument that private business is going to be the death of free speech is a strawman argument of the right. Typically this demographic goes out of its way to ensure a healthy free market with little obstructions. However they seem to lack the technical experience to launch a (successful and secure) competing social network of their own. So instead of achieving technological parity and becoming valid competition, they instead aim to impose regulations on the entire industry. It's ironic and disappointing.
- yurielt 6y ago
  >Typically this demographic goes out of its way to ensure a healthy free market with little obstructions. You are kidding right ? Private businesses goes out of their way to obstruct the market. Google being exemplary proof of this.
  - sjwalter 6y ago
    How does AWS caving to social pressure to fuck over Parler fit your narrative?
    - zelon88 6y ago
      If parler knew what the were up against they would have had on prem infra. Parler did not have technological parity with Facebook. It's embarrassing to say you're going to beat Facebook at their own game and then create a business model that's unsustainable from the very beginning.
      - sjwalter 6y ago
        Uh, wat? Parler's directly competing with Twitter. Why are you commenting when you have no idea what you're talking about?
- huntermeyer 6y ago
  Have you forgotten about Parler?
dmitryminkovsky 6y ago
> Why does it matter if the flagging is automatic? Does that somehow make it more acceptable? Neil Postman discusses this at length in Technopoly: The Surrender of Culture to Technology (1992): > Naturally, bureaucrats can be expected to embrace a technology that helps to create the illusion that decisions are not under their control. Because of its seeming intelligence and impartiality, a computer has an almost magical tendency to direct attention away from the people in charge of bureaucratic functions and toward itself, as if the computer were the true source of authority. A bureaucrat armed with a computer is the unacknowledged legislator of our age, and a terrible burden to bear.
- dkarl 6y ago
  AI is a new wrinkle in an old trick to distance elected officials from accountability. After the electrical blackouts last week in Texas, the weather-worthiness of power generation facilities was framed as a technical decision that engineers made, instead of a political decision to weigh costs and benefits of regulation. When the decision is framed this way, the public's attention is directed away from the elected officials who are supposed to be accountable, and for which we have created straightforward mechanisms of accountability. Public anger dissipates in a fog of uncertainty of how these unknown, anonymous engineers could be held accountable and what kind of accountability they should actually have. AI plays exactly the same role.
onion2k 6y ago
Why does it matter if the flagging is automatic? Does that somehow make it more acceptable? The problem is very simple. Unless social media companies either stop moderating content, which will kill their profits because advertisers will run away, or pay people to moderate everything, which will kill their profit because it'd be wildly expensive, they'll continue to do the half-assed version of moderating they do now using AI and actually make a profit but also make stupid mistakes like this. Computers and people don't need to operate perfectly. Expecting that is silly. So long as companies deal with errors reasonably it's all good.
- _qulr 6y ago
  > So long as companies deal with errors reasonably it's all good. We all eagerly await this utopia. Or maybe expecting that is silly, given the many years of empirical evidence?
  - qubex 6y ago
    There’s plenty of counterexamples one who might wish to dismiss the idea that technology can reasonably manage things, but all things considered, given the number of _potential_ instance of mishap that do not regularly occur any the relatively minor consequences of the mishaps that do occur, I’d conclude that on the whole arguing the original poster’s position is not inherently ‘_silly_’.
    - _qulr 6y ago
      > There’s plenty of counterexamples one who might wish to dismiss the idea that technology can reasonably manage things Are there? How many historical precedents are there for "curation" at the scale of today's tech giants: Facebook, Twitter, Google, Apple App Store, Amazon.com? IMO none of them do it well, or even reasonably. > given the number of _potential_ instance of mishap that do not regularly occur We only hear about the "notable" cases. Every day, people are accidentally or intentionally banned by Twitter or one of the other tech giants, and nobody ever hears about it, because these people are not famous and have nobody else to speak for them and raise a public fuss. > any the relatively minor consequences of the mishaps that do occur Regardless of one's opinion on the matter, you can hardly call the banning of the POTUS, for example, from the world's largest and most important social networks as a "relatively minor consequence". This is very consequential. You may think the consequences are good, or you may think they're bad, but it's consequential. If these "mishaps" are not consequential, why are we all here talking about them?
koonsolo 6y ago
Alright, let's say you get appointed at Twitter to clean up this mess. You get full authority to do whatever needs to be done. How would you solve this? Your solution will be criticized by the other HN-ers here :D.
- dumbfounder 6y ago
  It should be criticized, why do you imply that is a negative? AI is just a tool, and it can be used incorrectly. In this case we have a policy that seems to be applied incorrectly. That is a simple problem to solve, require more review, or the ability to escalate. Don't make such decisions "final".
  - koonsolo 6y ago
    Criticing is easy when you don't have to provide any solutions. Expecting Twitter not to have their hands full with these kinds of issues is very naive. It's always a question how to spend your very limited resources.
- pgrote 6y ago
  >Alright, let's say you get appointed at Twitter to clean up this mess. Announce a sale of intellectual and physical assets and close down. I would realize the company is a negative drain on society.
  - koonsolo 6y ago
    So basically some other platform will take over?
    - _qulr 6y ago
      No, break up all the tech giants, and prevent monopolistic platforms from ever arising. A platform so gigantic that it's impossible for humans to intelligently manage it should simply not exist.
jtdev 6y ago
It allows those responsible for these events to feign lack of agency. The Third Reich was enabled by convincing people that they had no agency and demanded that they adhere to the will of the bureaucracy - even when that meant genocide. Not saying that Twitter's censorship activities are comparable in terms of outcome, but rather that the same underlying mindset is at play here... of course the big tech censorship is likely to result in much more troubling outcomes in the future.
dragonwriter 6y ago
> Why does it matter if the flagging is automatic? Does that somehow make it more acceptable? Since the action (for “encouraging or promoting suicide”, bizarrely) was upheld on human review, whether the flagging was automatic or not is obviously irrelevant. Clearly, this action is unwarranted and the Bishop should be entitled to a full refund of monies paid to Twitter for the privilege of having messages distributed on the platform, at a minimum from the point of the unjustified ban.

thedanbob 6y ago

Or Gordon Dickson: https://www.atariarchives.org/bcc2/showpage.php?page=133 https://www.atariarchives.org/bcc2/showpage.php?page=133

teh_klev 6y ago
I remember reading this years ago, and it's still as bonkers as ever. Slightly related: https://en.wikipedia.org/wiki/Brazil_(1985_film) https://en.wikipedia.org/wiki/Brazil_(1985_film) "One day, shortly before Christmas, a fly becomes jammed in a teleprinter, misprinting a copy of an arrest warrant it was receiving resulting in the arrest and accidental death during interrogation of cobbler Archibald Buttle instead of renegade heating engineer and suspected terrorist Archibald Tuttle."
- 1234letshaveatw 6y ago
  how prescient that it was a bug :)
rzzzt 6y ago
"Do not fold, spindle or mutilate" was written on IBM punch cards.

forgingahead 6y ago

This is like the negative opposite of the happy coincidences that need to happen for a startup or new idealistic organisation to take off: The founder(s) have a crazy, innovative, idea to improve the world, and while they're perceived as mad by the masses initially, it's the first followers who give credence and legitimacy to this idea. So the first employees, initial customers, initial investors, and even just random supporters who independently believe in the value what the originators are doing, and that helps propel acceptance within the larger community and eventually the masses. This is the authoritarian opposite - really terrible decisions and mechanisms have been implemented and are being repeated on a daily basis by powerful companies or organisations, and while the masses find it worrying initially, there are apparently no shortage of apologists online and offline who are falling over themselves to helpfully explain to the rest of us why this crap is ok: "It's an AI algorithm" "They're a private company" "You don't have to use them" "Well, they're standing against bigotry" etc And then suddenly we find ourselves in an environment not unlike the Kafka-esque scenarios referenced. Be wary of "explainers" - often just they're consensus building, but it's not clear whether they're building in the right direction or the wrong one.

corty 6y ago

Funnily enough, Das Schloss is often interpreted to be a criticism of the church and organized religion in general. The castle and its workings are seen to be like the clerical teachings, arbitrary, incomprehensible and illogical, but nonetheless highly respected by the plebs.

throwaway0a5e 6y ago
How does the specific institution being ridiculed matter at all? Whether it's the Catholic church, Youtube or the DMV the point is it's a black box of procedural spaghetti that too often produces insane results and is seemingly unaccountable.
- corty 6y ago
  It doesn't really matter, but it is a funny coincidence that a cleric, as a representative of one kafkaesque institution is being censored by Twitter, another kafkaesque institution.
MarkPNeyer 6y ago
The role of the church 400 years ago is now similar to that played by the technocracy today.
keiferski 6y ago
Eh, I think Kafka was infinitely more subtle than that. Framing the story as a mundane criticism of the Church seems to be missing much of the depth of his writing. On that note, Kafka was also quite interested in Kabbalah, Jewish mysticism, which is almost inversely related to sociopolitical structures. https://en.wikipedia.org/wiki/Franz_Kafka_and_Judaism https://en.wikipedia.org/wiki/Franz_Kafka_and_Judaism Here's a great aphorism of his that's related to religion: “Leopards break into the temple and drink all the sacrificial vessels dry; it keeps happening; in the end, it can be calculated in advance and is incorporated into the ritual.”
lepton 6y ago
There was a time when the clergy was Catholic, but might various "scientific experts" be filling that role today?

guerrilla 6y ago

> future tyranny might end up not being Orwell's 1984 or Huxley's Brave New World, but a kafkesque nightmare where people are lost in a world of AI giving out absurd punishments. I think the present for some people is already a synthesis of all three and if we're not careful it will be for everyone.

ineedasername 6y ago

Automatic flagging just means their algorithm is crap. Which is bad, but it's a fixable technical problem in the medium term and the post can be restored in the short term. If this were deliberate flagging it would mean they were way outside of any reasonable interpretations of their guidelines and users really could no longer post anything and be confident it wouldn't run afoul of those interpretations. This is much harder to fix, and the confusion it wod cause is significantly higher.

nradov 6y ago
The technical problem isn't fixable without AGI.
ihsw 6y ago
The vagueness and ambiguity are features, not bugs -- it permits arbitrary punishment. Personally I'd rather they skip the appearance of reasoned arguments and go straight to "Twitter does not approve of this message." No muss, no fuss.
mdpopescu 6y ago
Ah... I'm sorry to say, ihsw, but you've been shadowbanned.

somehnrdr14726 6y ago

One difference is that Twitter, Facebook, Google, are not governments and have no real authority over you. They only have the power you give them by choosing to stay in their ecosystem. Therefore it's a strain to refer to them as officials. If you're worried about them _becoming_ officials, then act now. Leave these platforms and convince others to do the same.

abyssin 6y ago
We don't live in isolation, and most of us couldn't. If we want to be part of the social fabric, we're often forced to use these services to some extent, so they have real authority over us.
keiferski 6y ago
Well, in a literal sense, no. But they certainly have the ability to: - Exclude you from the largest (and nearly monopolizing) media channels. Good luck getting an online business off the ground when an AI filter bans you from YouTube, Facebook, Twitter, and Google search results. And good luck contacting Google support to contest it, which is notoriously absent for virtually everyone. - Have you socially ostracized by labelling you as X bad thing, all without you having any recourse or ability to contest the designation. The fact that stuff like this is legal blows my mind: https://www.cbsnews.com/news/yelp-racist-alert-added-business-review/ https://www.cbsnews.com/news/yelp-racist-alert-added-busines... - Collude to affect government policy and prevent competitors from gaining any footholds. Typical BigCo stuff.
- somehnrdr14726 6y ago
  Sticking to the consumer-side of the discussion, the BigCos aren't going to break the cycle. In fact, it's a capitalist crime to do so. Their flywheel requires a steady cattle population to surveil in the fields. When we choose as individuals to remain in their ecosystem, we choose the hidden fees of surveillance capitalism over more straightforward payment models. This is what gives BigCos the muscle to wield against smaller capitalists. We in our consumer role, by not participating in these business models, are the only truly free actors in the system at this time. The only other way out is for government to get bigger and step in, as it's doing in Australia now.

zelon88 6y ago

I think the argument that private business is going to be the death of free speech is a strawman argument of the right. Typically this demographic goes out of its way to ensure a healthy free market with little obstructions. However they seem to lack the technical experience to launch a (successful and secure) competing social network of their own. So instead of achieving technological parity and becoming valid competition, they instead aim to impose regulations on the entire industry. It's ironic and disappointing.

yurielt 6y ago
>Typically this demographic goes out of its way to ensure a healthy free market with little obstructions. You are kidding right ? Private businesses goes out of their way to obstruct the market. Google being exemplary proof of this.
- sjwalter 6y ago
  How does AWS caving to social pressure to fuck over Parler fit your narrative?
  - zelon88 6y ago
    If parler knew what the were up against they would have had on prem infra. Parler did not have technological parity with Facebook. It's embarrassing to say you're going to beat Facebook at their own game and then create a business model that's unsustainable from the very beginning.
    - sjwalter 6y ago
      Uh, wat? Parler's directly competing with Twitter. Why are you commenting when you have no idea what you're talking about?
huntermeyer 6y ago
Have you forgotten about Parler?

dmitryminkovsky 6y ago

> Why does it matter if the flagging is automatic? Does that somehow make it more acceptable? Neil Postman discusses this at length in Technopoly: The Surrender of Culture to Technology (1992): > Naturally, bureaucrats can be expected to embrace a technology that helps to create the illusion that decisions are not under their control. Because of its seeming intelligence and impartiality, a computer has an almost magical tendency to direct attention away from the people in charge of bureaucratic functions and toward itself, as if the computer were the true source of authority. A bureaucrat armed with a computer is the unacknowledged legislator of our age, and a terrible burden to bear.

dkarl 6y ago
AI is a new wrinkle in an old trick to distance elected officials from accountability. After the electrical blackouts last week in Texas, the weather-worthiness of power generation facilities was framed as a technical decision that engineers made, instead of a political decision to weigh costs and benefits of regulation. When the decision is framed this way, the public's attention is directed away from the elected officials who are supposed to be accountable, and for which we have created straightforward mechanisms of accountability. Public anger dissipates in a fog of uncertainty of how these unknown, anonymous engineers could be held accountable and what kind of accountability they should actually have. AI plays exactly the same role.

onion2k 6y ago

Why does it matter if the flagging is automatic? Does that somehow make it more acceptable? The problem is very simple. Unless social media companies either stop moderating content, which will kill their profits because advertisers will run away, or pay people to moderate everything, which will kill their profit because it'd be wildly expensive, they'll continue to do the half-assed version of moderating they do now using AI and actually make a profit but also make stupid mistakes like this. Computers and people don't need to operate perfectly. Expecting that is silly. So long as companies deal with errors reasonably it's all good.

_qulr 6y ago
> So long as companies deal with errors reasonably it's all good. We all eagerly await this utopia. Or maybe expecting that is silly, given the many years of empirical evidence?
- qubex 6y ago
  There’s plenty of counterexamples one who might wish to dismiss the idea that technology can reasonably manage things, but all things considered, given the number of _potential_ instance of mishap that do not regularly occur any the relatively minor consequences of the mishaps that do occur, I’d conclude that on the whole arguing the original poster’s position is not inherently ‘_silly_’.
  - _qulr 6y ago
    > There’s plenty of counterexamples one who might wish to dismiss the idea that technology can reasonably manage things Are there? How many historical precedents are there for "curation" at the scale of today's tech giants: Facebook, Twitter, Google, Apple App Store, Amazon.com? IMO none of them do it well, or even reasonably. > given the number of _potential_ instance of mishap that do not regularly occur We only hear about the "notable" cases. Every day, people are accidentally or intentionally banned by Twitter or one of the other tech giants, and nobody ever hears about it, because these people are not famous and have nobody else to speak for them and raise a public fuss. > any the relatively minor consequences of the mishaps that do occur Regardless of one's opinion on the matter, you can hardly call the banning of the POTUS, for example, from the world's largest and most important social networks as a "relatively minor consequence". This is very consequential. You may think the consequences are good, or you may think they're bad, but it's consequential. If these "mishaps" are not consequential, why are we all here talking about them?

koonsolo 6y ago

Alright, let's say you get appointed at Twitter to clean up this mess. You get full authority to do whatever needs to be done. How would you solve this? Your solution will be criticized by the other HN-ers here :D.

dumbfounder 6y ago
It should be criticized, why do you imply that is a negative? AI is just a tool, and it can be used incorrectly. In this case we have a policy that seems to be applied incorrectly. That is a simple problem to solve, require more review, or the ability to escalate. Don't make such decisions "final".
- koonsolo 6y ago
  Criticing is easy when you don't have to provide any solutions. Expecting Twitter not to have their hands full with these kinds of issues is very naive. It's always a question how to spend your very limited resources.
pgrote 6y ago
>Alright, let's say you get appointed at Twitter to clean up this mess. Announce a sale of intellectual and physical assets and close down. I would realize the company is a negative drain on society.
- koonsolo 6y ago
  So basically some other platform will take over?
  - _qulr 6y ago
    No, break up all the tech giants, and prevent monopolistic platforms from ever arising. A platform so gigantic that it's impossible for humans to intelligently manage it should simply not exist.

jtdev 6y ago

It allows those responsible for these events to feign lack of agency. The Third Reich was enabled by convincing people that they had no agency and demanded that they adhere to the will of the bureaucracy - even when that meant genocide. Not saying that Twitter's censorship activities are comparable in terms of outcome, but rather that the same underlying mindset is at play here... of course the big tech censorship is likely to result in much more troubling outcomes in the future.

dragonwriter 6y ago

> Why does it matter if the flagging is automatic? Does that somehow make it more acceptable? Since the action (for “encouraging or promoting suicide”, bizarrely) was upheld on human review, whether the flagging was automatic or not is obviously irrelevant. Clearly, this action is unwarranted and the Bishop should be entitled to a full refund of monies paid to Twitter for the privilege of having messages distributed on the platform, at a minimum from the point of the unjustified ban.