imagine if it does it based on nothing but other players reviews.

https://imgur.com/PqPwIhr

https://imgur.com/lVOt4P3

https://imgur.com/WmEDR31

Originally posted by eram:

imagine if it does it based on nothing but other players reviews.

https://imgur.com/PqPwIhr

https://imgur.com/lVOt4P3

https://imgur.com/WmEDR31

It could have section for Troll best-off :p

Ai is no where near good enough to work out the difference. it will end up telling everyone press K to kiss your friends on the mouth

Originally posted by eram:

Ai is no where near good enough to work out the difference. it will end up telling everyone press K to kiss your friends on the mouth

O I think it's good enough and CHatGPT below isn't fine tuned for that:

Based on an analysis of Steam reviews and online discussions, Dragon's Dogma 2 has elicited mixed reactions from players. Here is a breakdown of the common likes and dislikes, along with their respective frequencies:

Likes:
Combat System (frequently praised):

The dynamic and engaging combat system, especially with interactions like climbing large enemies or combining unique pawn abilities, is a standout feature. Players enjoy the variety in classes and the nuanced approach to battles involving large monsters like chimeras and griffins.
Pawns System (often highlighted):

The AI-controlled companions called "pawns" are celebrated for their functionality and how they add depth to gameplay by adapting to situations and player commands.
Monster Encounters (frequently noted):

The thrill of encountering massive monsters in unexpected scenarios, such as a griffin attack during another battle, adds unpredictability and excitement.
Immersive World (frequently appreciated):

Players enjoy the world’s detail, including environmental interactions, physical systems, and its blend of medieval fantasy aesthetics.
Dislikes:
Microtransactions (very frequently criticized):

Many reviews complain about unnecessary microtransactions for items that can be acquired in-game, such as cosmetic changes or in-game utilities, viewing these as exploitative.
Performance Issues (often noted):

Complaints about poor optimization, including frame rate dips and bugs, are common. These issues are more pronounced for players with mid-range hardware setups.
Archaic Design Choices (frequently criticized):

Some aspects, like confusing quest markers, poorly explained objectives, and limited fast travel options, feel outdated and detract from modern expectations.
Recycled Elements (occasionally mentioned):

While fans of the original appreciate familiarity, others feel that the sequel doesn’t innovate enough and retains outdated mechanics from the 2012 original.
Narrative and Quest Issues (less frequent but noted):

Poor quest logic, broken objectives, and unclear instructions lead to player frustration, especially in settlement-based missions.
The game currently holds a "Mostly Negative" rating on Steam due to these issues, with the microtransactions and performance problems being the most cited grievances. However, many players highlight that the core gameplay remains enjoyable and has the potential to improve with patches and updates. With time, it may follow other games that rebounded after addressing launch issues.

For a more detailed dive into community sentiment, you can explore reviews and discussions on platforms like Steam and PC gaming forums.

it isnt good enough, proof, all real world examples so far.

Originally posted by eram:

it isnt good enough, proof, all real world examples so far.

I think the one I did with Dragon Dogma is pretty on point

are you saying you used ai to write a review for a game for you?

Originally posted by eram:

are you saying you used ai to write a review for a game for you?

No
Sorry I though it was beyond obvious but I guess you didn't even read it

I ask chat GPT to do what I proposed for Steam

I ask it: "could you make a summary of the steam reviews for the game dragon dogma 2 listing with their respective frenquency what people like about the game and dislike?"

The result is in my previous post and it is really good disproving your comment "it isnt good enough, proof, all real world examples so far."

oh you just mean that post not a review, yeah its ai slop
a human will be better at reviewing a game they played

ai summary based on players reviews will produce more slop, trolling, fake info etc.

Originally posted by eram:

oh you just mean that post not a review, yeah its ai slop
a human will be better at reviewing a game they played

ai summary based on players reviews will produce more slop, trolling, fake info etc.

Ok you seem to have a lot of preconceived opinions about it.
It will probably be more productive for me talking to my wall I guess.

Originally posted by Pnume:

Originally posted by eram:

oh you just mean that post not a review, yeah its ai slop
a human will be better at reviewing a game they played

ai summary based on players reviews will produce more slop, trolling, fake info etc.

Ok you seem to have a lot of preconceived opinions about it.
It will probably be more productive for me talking to my wall I guess.

You ignoring just how bad "AI" is right now doesn't actually help your suggestion.

"AI" is still trying to tell users to kill themselves or leave their SOs for them and hallucinates other dangerous stuff...

Mother sues AI chatbot company Character.AI, Google over son's suicide

https://www.ctvnews.ca/sci-tech/mother-sues-ai-chatbot-company-character-ai-google-over-son-s-suicide-1.7087274

Man ends his life after an AI chatbot 'encouraged' him to sacrifice himself to stop climate change

https://www.euronews.com/next/2023/03/31/man-ends-his-life-after-an-ai-chatbot-encouraged-him-to-sacrifice-himself-to-stop-climate-

Gemini is supposed to have restrictions that stop it from encouraging or enabling dangerous activities, including suicide, but somehow, it still managed to tell one "thoroughly freaked out" user to "please die".

https://news.sky.com/story/googles-ai-chatbot-gemini-tells-user-to-please-die-and-you-are-a-waste-of-time-and-resources-13256734

ChatGPT keeps hallucinating—and that's bad for your privacy

https://www.techradar.com/computing/cyber-security/chatgpt-keeps-hallucinatingand-thats-bad-for-your-privacy

no "AI" is not ready and what you refer to as "AI" is not actually AI. We don't even have actual AI right now.

We have whats called Large Language Models. It will use data for EVERY post. It can not tell whats a troll post or a sarcastic post or a farming post or anything like that. It will take data from ALL of them and pump out junk.

This about this for a second... if google, a HUGE company who has billions of dollars and thousands of programmers, who is in the information tech industry, who has been working on AI/Large Language Models for decades (yes they have been worked on for that long or even longer), can't get their "AI" bots to behave and not encourage people to self harm what makes you think Valve, a gaming company of only 350ish people, many of which are not programmers, could make a better "AI"?

If you want to know how large language models such as chatgpt and google gemini work, here's a short explainer video for beginners:

https://www.youtube.com/watch?v=LPZh9BOjkQs

There's also a longer course on the same channel if you want to learn more.

To summarize something that is already a summary, no, LLMs don't "think". They simply generate a list of the possible words that might come next, and then pick one of the likely words at random.

Ooooh, can we use it to seperate the reviews about "woke", "anti-woke", "Denuvo", "CEO kicked a puppy" or whatever it is that gets peoples panties in a twist into a category that automatically gets hidden?

I could see value in that. I really could.

As long as it can be disabled, i have no interest in reading AI generated word vomits.