Path: ...!weretis.net!feeder9.news.weretis.net!news.quux.org!eternal-september.org!feeder3.eternal-september.org!news.eternal-september.org!eternal-september.org!.POSTED!not-for-mail
From: Your Name <YourName@YourISP.com>
Newsgroups: misc.phone.mobile.iphone
Subject: Re: iPhone dictation bug
Date: Thu, 27 Feb 2025 15:54:35 +1300
Organization: A noiseless patient Spider
Lines: 41
Message-ID: <vpok5a$2r3co$1@dont-email.me>
References: <320911a8f6ef9f8fa762732d3ae5eec398d1ddc7@i2pn2.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: 8bit
Injection-Date: Thu, 27 Feb 2025 03:54:36 +0100 (CET)
Injection-Info: dont-email.me; posting-host="08427bbace0abc35fa660c1418dc221d";
	logging-data="2985368"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX1++J56F8mnF+4geQnkPciGpWpaHOhz1nuA="
User-Agent: Unison/2.2
Cancel-Lock: sha1:iP+yb5uSGOXucGUlhaU/BGY/Npw=
Bytes: 2833

On 2025-02-27 02:27:34 +0000, Rick said:
> On 2/26/2025 9:12 PM, Alan wrote:
>> On 2025-02-26 17:32, badgolferman wrote:
>>> Apple is fixing a bug within the dictation feature on some iPhones that
>>> briefly suggests the word “Trump” when a word with an R consonant is
>>> spoken, including “racist.”
>>> 
>>> The company is responding to the controversy after some iPhone owners
>>> posted videos on social media this week to detail how the glitch works.
>>> 
>>> When users activated the dictation feature and said the word “racist,” the
>>> word “Trump” appears in the text window before quickly being replaced by
>>> the correct word, according to various videos posted online.
>>> 
>>> “We are aware of an issue with the speech recognition model that powers
>>> Dictation and we are rolling out a fix today,” Apple said in a statement
>>> sent to The Associated Press on Wednesday.
>>> 
>>> https://apnews.com/article/apple-iphone-racist-trump-glitch-d80f88d69f6ceac585904f2faa2a9212 
>>> 
>> 
>> Sounds like someone decided to have a little fun.
>> 
>> Good.
>> 
>> :-)
> 
> I'd hardly call it fun when someone puts a feature like this into 
> production software.  If the person who did this can be identified, 
> they should be fired and possibly even prosecuted.  Imagine the outcry 
> if it had flashed the word "Obama"?

It's unlikely to be an actual bug nor a deliberate act. There are many 
many places were the words "Trump" and "racist" appear, so useless AI 
bots simply connect the two and offer it as a suggestion. Now a 
specific rule will have to be added to the dictation feature to avoid 
that connection. Welcome to the world of idiotic AI and its garbage 
answers. There will be a ton of similar issues that simply haven't been 
stumbled across.