HN Reader

Timeless quote:

> If Jimmy Allred says it’s raining, and W. Lee O’Daniel says it isn’t raining, Texas newspapermen quote them both, and don’t look out the window to see which is lying, and to tell the readers what the truth is at the moment.

https://quoteinvestigator.com/2023/11/14/rain-look/

If an AI tool, or for that matter a meddling editor, says a headline is "framing the action as unprecedented in a way that might subtly critique the administration", the correct response is "yes, that was the idea".

The problem with trying too hard to neutrally report both sides of a story is that's not unbiased either, it's just biased in favour of the side that can lie the most convincingly.

> Several sentences in the story were flagged as biased, including this one: "It’s the first time in 60 years that a president has mobilized a state’s National Guard without receiving a request to do so from the state’s governor." According to the bias indicator, this sentence is "framing the action as unprecedented in a way that might subtly critique the administration." It was best to give more context to “balance the tone."

"Might subtly critique"?! This isn't protecting versus bias, this isn't even charitably regardable as cowardice.

It's an attack on reason & reporting. To shield the worst cry bullies & offenders. It's ghastly how basic simple reporting so offends, that the public interest is sabotaged by these foes. We face such a need to improve the media, and this is the opposite path, of craven blunt un-intelligently intermediated corporatism.

I said to a colleague the other day that Law360’s reporting seems to be more tortuous and weird than usual. I wonder if this is why.

There was one piece on a case I’ve been following quite closely where I was genuinely unable to make sense of what had happened from the Law360 report without sitting down and spelling my way through it.

Remember kids, if you don't like the truth, just accuse it of being biased and force it to change!

LLMs and ML algorithms are beginning to influence the entire lifecycle of articles: researching, writing, editing, publication, discovery (TikTok), and consumption (ChatGPT summarize this). With the few big players, it could be the same model involved at every step. It's scary how a small change to a system prompt could subtly influence things across the board and guide popular opinion.

Who checks the bias detector for bias?

I have so many question.

Let's say a media outlet, like a newspaper, commits to common journalism best practices, such as neutrality. In my opinion, this does not mean avoiding certain words; it means presenting different perspectives on a topic based on evidence.

Besides that, if someone writes an Op-Ed or commentary, the author expresses their personal opinion. How does this work with this AI thingy? Currently it looks like plain censorship, isn't it?

And why should I buy a newspaper that does not represent my political or personal opinion? I thought that was how it worked.

To me, a newsroom's bias IS its value. If you let me know your values, I can actually interpret what you're saying. It's the pretense of impartiality that makes me distrust someone.

Does this mean automated censorship by AI "bias" detection?

Who watches the watchers?

Today, somewhere in the world, some people made an event occur. Some people approved and some people disapproved. It impacted some people more than others, which some people think may be a bad thing, although others think that it may have been a good thing. The event stemmed from many past events, which many people have various strong opinions about. Politicians made various strong statements about the event, as well as a few celebrities. The future impact of the event is under debate, though experts agree that the situation is complicated and people should probably think about it very carefully before coming to conclusions. It is unclear what the market will do in response to this event.

"Bias" and "narrative" have become meaningless words that people use when encountering ideas they don't like. Or when they believe that neither side could ever be right about an issue, that it's virtuous for a stance to be in the middle of the current Overton window.

What I do not understand is this top-down mandate that people use specific tools.

Mandate productivity, mandate quality; the tools will be adopted if they are sharp.

But elevating the way work is performed above the work itself? Weird.