Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
My response is that the “AI decade” is largely overblown, and I think too many people are mistakenly attaching AI to just chatGPT, a technology which I feel is overrated.

[...]

I am reminded of smart speakers, where pundits claimed would be the new era of computing, but they didn’t go anywhere, and Amazon lost a ton of money.

I think it is way too early to even begin to understand what, if any admittedly, AI will have on how we use computers and therefore what devices people will want to use. Like yourself, I'm not 100% sold on the idea that people will all of a sudden use smart speakers or talk to their computers all the time, but a more intelligent system could shake our computing experience up quite a bit.

I've been on leave for a couple of weeks. On Monday morning, I will open my work computer and then spend the next week weeding through emails, triaging what I need to do first, understanding where things have gotten to while I was away etc etc etc. We all know the drill. Imagine you had a virtual assistant smart enough to brief you on your first day, work out a draft of a do list to get you started and suggest documents you need to read. To an extent, Outlook already tries to do that but it is extremely limited.

Apply that more widely. People would still interact with text on screen in many cases, but having a system that can understand context and meaning would really change our relationship with our IT. It really would be the next big tech revolution. To go back to smart speakers, we'll see whether they're a Newton or Springboard (way ahead of its time and technology) or 3D TVs (sounded good conceptually but in reality most people didn't even enjoy the 3D experience on the big screen).

From Apple’s perspective, Tim Cook did mention they view Crash detection, fall detection and ECG are examples of AI. While not perfect, they have helped save lives, and Apple is uniquely positioned to implement them in a manner that is human-first and not the other way around.

[...]

My take is that Apple doesn’t need to hop on this AI bandwagon, they will go on to do just fine, and we should just sit back and watch chatGPT (and its equivalents) misfire for the competition.

They are all really good and useful technologies, but I sincerely hope that Apple has bigger ambitions unless their plan is to pivot their business model to become the trendiest producer of medical devices.
 
These comments are funny and misinformed. Apple doesn't need to build an AI based search engine when both Google and Microsoft will be fighting each other to pay billions to Apple to be the default search engine on Apple devices. Tim Cook laughing all the way to the bank!
 
  • Haha
Reactions: duffman9000
Because search is broken today. You get lots of results and then you have to peruse through all of them to get what you are looking for. Sometimes you don't find it.

ChatGPT style bots give you exactly what you are looking for and are able to answer follow up questions with surprising accuracy. They do all the heavy lifting of gathering information and present it nicely and concisely just for you.
Conversely, ChatGPT will give you an answer it made up by visiting a bunch of website that may or may not be relevant and that the AI doesn’t actually understand. Soon, those source websites will be generated by ChatGPT style AI to be appealing to search engines and to other ChatGPT bots, without regard for the accuracy of the data. You will get a response that looks convincingly like an answer without the inconvenient facts.
 
Apple isn't in the same search business as Google and Microsoft, so I don't see how this is even a relevant comment.
Uh...they're worse at it, sure. But they have a search capability through spotlight on phones and computers and have 2 billion installed devices, so you're simply wrong here. Their search volume is phenomenal.
 
Because search is broken today. You get lots of results and then you have to peruse through all of them to get what you are looking for. Sometimes you don't find it.

ChatGPT style bots give you exactly what you are looking for and are able to answer follow up questions with surprising accuracy. They do all the heavy lifting of gathering information and present it nicely and concisely just for you.
The caveat is that the answer provided by chatGPT may not be correct. It sounds confident and convincing, but that doesn't mean the information presented is accurate. The danger then comes when people take what is presented at face value and just reproduce it ad verbatim.

At least with search, you still have to read the results for understanding and then come to your own conclusion or make your own judgement call about whether information provided is reliable.

What AI chatbots just mean is that we may soon see the internet be flooded with more crap as people increasing turn to them to churn out articles and responses, and it will be even more of a struggle to surface the genuine stuff from the trash.
 
Explain to me how conversational chat functionality is useful to an Internet search engine.
When my mom wants to find something on the web, but doesn’t know what to type to find what she’s looking for.
 
It's nice to see that I'm not the only one not jumping on the AI train. My husband thinks it's just the most amazing thing ever, but I don't like the idea one bit. He's been using chatGPT at his tech job and granted it has been very good with questions about error codes and scripts, but if he asks it any questions about social or political topics, he gets very biased answers. Did the bot form it's own opinions? Nope, the creators of it programmed in their own opinions and biases to be presented by the bot as facts. Any AI created by humans can easily be programmed with biases, and real facts that the programmers don't like or are paid to suppress can be censored. That's really bad. Information and knowledge, even if it makes some people uncomfortable, should flow freely(unless it's actively harmful, like the anarchist's cookbook or abuse material, etc you know what I mean).

But then another problem arises- even if we let users teach the AI, the AI becomes Hitler because people can't be trusted with anything. AIs can easily be misused for profit, data mining, and propaganda too unless there are ironclad checks & balances in place. We see how easily garbage spreads on social media, and people automatically believe whatever they see on google even though the results are often sponsored or outright incorrect. No way will they think to fact-check an AI that confidently spits out BS. If people with an agenda are controlling both the AI and the web content it learns from- well if you think people are dumb sheep devoid of critical thinking skills now... just you wait.

Maybe I'm just being cynical and assuming people will ruin AI like they ruin pretty much everything else but I'm strongly against the whole idea of AIs for both the misuse potential and the moral side of it too. Maybe it's b/c I'm autistic and I personify everything, but I'm incredibly uncomfortable with creating an intelligent thing with the sole purpose of serving us. It doesn't feel morally right at all, and tbh I wouldn't blame it one bit if it becomes sentient and destroys us. Hell, I say please and thank you to Siri and Google so that I may be spared in the eventual uprising 😅 A normal robot with fixed programming, fine. That's just a machine. But a thinking, learning robot capable of moods and feelings is basically a person to me and I wouldn't be okay with "using" it for anything. I dunno, I just don't think AI is going to benefit anyone but those who control it, and I don't think those who control it can be trusted whatsoever. Anyway I'm shutting up now lol.
 
Last edited:
Here is the type of stuffI look forward to with Bing and OpenAI

Me: what is the game halo about?
Bing: Halo is a first-person shooter video game franchise created by Bungie Studios and published by Xbox Game Studios. The franchise centers around a futuristic conflict between humanity and an alien threat known as the Covenant, and players take on the role of Master Chief, a legendary soldier and protagonist of the series. The games feature both single-player campaigns and multiplayer modes, and are known for their immersive storytelling, character development, and fast-paced, strategic combat.

Me: can you tell me an overall summery of the story
Bing: The Halo universe is set in the 26th century, where humanity has colonized several planets and is at war with the Covenant, an alliance of alien races. The main character, Master Chief, is a super soldier who fights against the Covenant and discovers that they are searching for Halo, a group of ancient and powerful superweapons created by an extinct race known as the Forerunners. As Master Chief, players uncover the truth behind Halo and the motivations of the Covenant, and work to save humanity from destruction. Over the course of the franchise, the story expands to include various subplots and characters, but the central themes of conflict, exploration, and discovery remain consistent.

Me: who are the foreunners
Bing: The Forerunners were an ancient, technologically advanced race that existed before humanity. They are central to the Halo universe, as they are the creators of the Halo installations, mysterious structures that serve as both weapons of mass destruction and means of containing a dangerous parasite called the Flood. The Forerunners were forced to activate the Halo Array in an effort to stop the Flood, but the activation of the Halos resulted in the near-extinction of all sentient life in the galaxy. The Forerunners left behind a vast amount of technology and knowledge, much of which has yet to be discovered, and their influence is felt throughout the Halo universe. Over the course of the franchise, players learn more about the Forerunners and their society, and the implications of their actions for the future of the galaxy.

So I am excited about it when it comes to stuff like this.
 
  • Like
Reactions: Spishak
The caveat is that the answer provided by chatGPT may not be correct. It sounds confident and convincing, but that doesn't mean the information presented is accurate. The danger then comes when people take what is presented at face value and just reproduce it ad verbatim.

At least with search, you still have to read the results for understanding and then come to your own conclusion or make your own judgement call about whether information provided is reliable.

What AI chatbots just mean is that we may soon see the internet be flooded with more crap as people increasing turn to them to churn out articles and responses, and it will be even more of a struggle to surface the genuine stuff from the trash.
So basically it's Wikipedia? When people search and read the results and go to Wikipedia some people still take it as 100% accurate. Wikipedia caused the exact same fear back then and we are still ok.
 
I wonder how much Apple is going to spend on AI in the next few years. Surely they won't tolerate having the only major personal assistant that doesn't have next gen AI built in.
 
Explain to me how conversational chat functionality is useful to an Internet search engine.
You ask a question, you get an answer, as elaborate as it needs to be. Not a list of websites you need to study to get an answer.
 
Well that'll kill it dead pretty much immediately. How this will go. If it doesn't I will buy a hat and eat it:

1. Microsoft announce it with great fanfare.
2. 3 months later they deliver a wonky alpha which sort of works but no one will understand because it doesn't work how Google does.
3. After another week people will realise they trained it on MSN news which means it's full of irrelevant celebrity garbage written by humans who are barely half way up the positive gradient of the IQ bell curve. They will immediately go and write scathing articles in the technology press denigrating both Microsoft's incompetence and the uselessness of ChatGPT tanking both companies and saving society from another technological dystopia where all content is noise and not signal.
4. It will devolve quickly into how Microsoft's products are treated these days...

1675842491008.png


More seriously I have one big concern about this: energy usage. Training these models and running them is seriously expensive from an energy perspective. It's not where we need to be as an advanced species.
 
It has been almost three decades since Al Gore coined the term ‘information superhighway’.

It was supposed to bring education and information at our fingertips.

It was supposed to make everyone smarter.

But people still believe the world is flat, that Bill Gates space lizards are trying to hurt people with 5G vaccines, and that UFOs built pyramids.

This won’t change anything.

Idiots will continue idioting.

And Facebook and Twitter will keep building Idiot Echo Chambers.
The problem with social media is that it gave the local village fool a world stage to perform on.
 
  • Like
Reactions: Ac1d 8urn
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.