January rumours

All advertisments are hidden for logged in members, why not log in/register?




I believe that some of our new loan players we will get will be in the list pictured - players not registered in the PL 25 man squad, however are eligible to play as born on or after 1st Jan 2004 (Under 21 player) Ignore Southampton as this list includes teams who play in the PL U21 top division as well

View attachment 228095

If you've pulled this from AI you should say so up front so that people know it's dubious and not to be trusted.
 
Yes I pulled the data from Google AI...not sure why it shouldn't be trusted but hay ho noted

I wouldn't necessarily say it shouldn't be trusted but I think you should say you got it from AI.

Ironically I got recently accused of using AI after I'd sat bloody listening to one of CW's pressers for half an hour and transcribed a summary of the main points 🤣
 
Suppose I should apologise for an off-topic comment. Anyone looking for more rumours, feel free to skip this.


Yes I pulled the data from Google AI...not sure why it shouldn't be trusted but hay ho noted

It's because all generative AI (Gemini, ChatGPT, Grok, etc.) is just a pattern-matching game. They're putting together the words that seem to fit best in response to your question, one after another. They're not reading your question, understanding it and looking up an accurate answer.

They have been specifically engineered to sound persuasive, and the technology is now good enough that they usually do so very well. Which is a problem because when we read something that sounds really credible it’s easy to forget that on a fundamental level they are not.

The algorithms are becoming more accurate as well, especially now they can access the live internet, but that doesn't mean that each response is drawn from accurate sources (this was somewhat easier to understand back when they were first released because they'd been trained only on historic data, so answers about current events might be very persuasive but couldn't possibly be correct. These days the source accuracy is much less evident). So for instance given your prompt the response might have replicated the structure of a similar reply about Arsenal's squad players, and then filled in fake names for the other teams because it didn’t reference texts which included them.

There's also the issue of hallucination, where an AI bot is provided with all the correct data but still corrupts the answer in a way which makes it inaccurate for reasons that, frankly, nobody is entirely clear about (which in itself is an indication of why we should be very cautious around AI). The other day I gave Gemini a photograph of some books and asked it for the authors and titles, and it gave a great response - for the first half, then confidently finished up with a load of books that didn't exist.


It's often the case that the responses from generative AI are accurate, but due to the issues mentioned above you have to sense-check every single point of data to have confidence in the output. A number of people on here have made themselves look daft this year posting inaccurate AI stuff on here because they failed to do this.

One thing you can do is ask AI for its source, check the source is credible and then share the source instead. Alternatively at a bare minimum please at least tag it as AI.


At the end of the day, generative AI can be useful for creative tasks (like brainstorming ideas) or where you can directly check the entire output (like simple coding tasks which either work or don’t). But due to the way they’re designed you should never trust them for sourcing facts or for providing analysis.
 
Suppose I should apologise for an off-topic comment. Anyone looking for more rumours, feel free to skip this.




It's because all generative AI (Gemini, ChatGPT, Grok, etc.) is just a pattern-matching game. They're putting together the words that seem to fit best in response to your question, one after another. They're not reading your question, understanding it and looking up an accurate answer.

They have been specifically engineered to sound persuasive, and the technology is now good enough that they usually do so very well. Which is a problem because when we read something that sounds really credible it’s easy to forget that on a fundamental level they are not.

The algorithms are becoming more accurate as well, especially now they can access the live internet, but that doesn't mean that each response is drawn from accurate sources (this was somewhat easier to understand back when they were first released because they'd been trained only on historic data, so answers about current events might be very persuasive but couldn't possibly be correct. These days the source accuracy is much less evident). So for instance given your prompt the response might have replicated the structure of a similar reply about Arsenal's squad players, and then filled in fake names for the other teams because it didn’t reference texts which included them.

There's also the issue of hallucination, where an AI bot is provided with all the correct data but still corrupts the answer in a way which makes it inaccurate for reasons that, frankly, nobody is entirely clear about (which in itself is an indication of why we should be very cautious around AI). The other day I gave Gemini a photograph of some books and asked it for the authors and titles, and it gave a great response - for the first half, then confidently finished up with a load of books that didn't exist.


It's often the case that the responses from generative AI are accurate, but due to the issues mentioned above you have to sense-check every single point of data to have confidence in the output. A number of people on here have made themselves look daft this year posting inaccurate AI stuff on here because they failed to do this.

One thing you can do is ask AI for its source, check the source is credible and then share the source instead. Alternatively at a bare minimum please at least tag it as AI.


At the end of the day, generative AI can be useful for creative tasks (like brainstorming ideas) or where you can directly check the entire output (like simple coding tasks which either work or don’t). But due to the way they’re designed you should never trust them for sourcing facts or for providing analysis.
You got that off AI dint you
 
How does this work
We already have one of their centre backs on loan
Do you mean loaning two? We did the same with Doyle and McAtee and JRS both an issue in the Chanpionship.

Or do you mean the fact we are potentially loaning two players who operate in the same position? I.e. they’d be potentially cutting each other’s lunch for a starting spot?
 
Do you mean loaning two? We did the same with Doyle and McAtee and JRS both an issue in the Chanpionship.

Or do you mean the fact we are potentially loaning two players who operate in the same position? I.e. they’d be potentially cutting each other’s lunch for a starting spot?
The latter
 
The latter
Agree seems an odd one. Though Dyche probably doesn’t mind helping his old mate Wilder, it does seem counterintuitive for the players. Or maybe Wilder feels Bindon simply hasn’t done enough and sees Boly and Mee as better bets alongside Tanganga?

We’ll find out in the next few weeks anyway!
 
35 next month, contract until the end of the season would be fine. Can’t say I know much about him. Obviously I’ve heard of him but is he no nonsense or a bit more of a ball player?
 
Wily Boly is on a free in the Summer, might just be a case of taking over his contract until the end of the season, so not taking up a loan spot.

Interesting that Oxford United accidentally leaked their transfer targets on X over the weekend.
I think Matos and (surprisingly) Bindon were on there. There might be nothing in it, but maybe he's been told we might be sending him back.
 
I think Bindon has done enough to be at least first choice back up, if not a starter, to Mee.

I like Mee but if there's a chance to keep Bindon permanantly then he could feasibly become a bone-fide starter for the next few years. He's much, much better than McGuiness and Mee only has a season or so left in him at this level.
 
I think Bindon has done enough to be at least first choice back up, if not a starter, to Mee.

I like Mee but if there's a chance to keep Bindon permanantly then he could feasibly become a bone-fide starter for the next few years. He's much, much better than McGuiness and Mee only has a season or so left in him at this level.
and Mee didn't exactly look great either.
 



Brentford seem to be in a decent position in the Premier League, so maybe they'd let Ethan Pinnock out on loan. 32 years old, he's mainly a bench player with only 3 appearances this season.
Left footed, 6ft 4" and capable of the odd goal.
Tanganga and Pinnock would be pretty decent partnership.
He'll be up in Sheffield this weekend as well.
 
Brentford seem to be in a decent position in the Premier League, so maybe they'd let Ethan Pinnock out on loan. 32 years old, he's mainly a bench player with only 3 appearances this season.
Left footed, 6ft 4" and capable of the odd goal.
Tanganga and Pinnock would be pretty decent partnership.
He'll be up in Sheffield this weekend as well.
I'm liking this option, just hope it's not as a makeweight for Peck going the other way.

A central defensive choice of four from Tanganga, Pinnock, Bindon and Mee is something I can get on board with.

McGuinness can be let go if anyone will have him though I don't doubt that we've probably overpaid for him and are stuck with him.

Get Zatterstrom out on loan and if he's any good can be phased in as a genuine back up when Mee leaves/retires - he seems to be doing a fair bit of media stuff so its not like he's not got something lined up to do in the next 18 months or so.
 
Facebook posts randomly up about signing Marc Leonard from Birmingham
 
Facebook posts randomly up about signing Marc Leonard from Birmingham
Saw that but cannot see where he fits in as he is a central midfield player, who is not the type to impose himself on the game.
I would have expected us to sign a wide man rather than his “type”
 
Facebook posts randomly up about signing Marc Leonard from Birmingham
Comes from a journo on Twitter. Having read the article, it doesn't say we're in for him as such. more that we've showed an interest in the past.

 
35 next month, contract until the end of the season would be fine. Can’t say I know much about him. Obviously I’ve heard of him but is he no nonsense or a bit more of a ball player?
Wily Boly is more of a side ways mover. A song and a dance over nothing is how I see it. ;)
 
Following on from Cassius Kray's very sensible and timely reminder about the limitations and dangers of trusting AI for sourcing facts or for providing analysis (message #66 above), it reminded me of the tactical analysis I saw on the match thread of our recent home game against Birmingham City on one of their forums . . .


To save you having to soil yourselves on their grubby forum, here's what it said . . .

AI Response to "How to beat Sheffield United at home" - these summaries haven't been too bad, but are a bit generic, however some people really like them

To beat Sheffield United, exploit their aggressive overlapping full-backs by overloading the left flank they often attack from, use quick transitions to catch them high up, and maintain a strong defensive shape to counter their possession-based short passing, especially disrupting ball-playing CBs like Jack O'Connell who start attacks.

Key tactics involve isolating their aggressive right-sided players (Baldock/Basham) and stretching their back three with swift attacks down the wings to create overloads or exploit space.
Key Tactical Approaches

Target the Left Flank (Their Strength):
Under Chris Wilder, they often overloaded the left with overlapping centre-backs (like O'Connell) and wing-backs.
Counter: Use quick, direct play down that side to isolate their defenders or switch play quickly to the right when they commit players forward.

Disrupt Their Build-Up:
Sheffield United (especially in past successful eras) used ball-playing centre-backs to initiate attacks.
Counter: Press their defenders to force errors, particularly the left-sided centre-back, to stop attacks starting from the back.

Exploit Transitions:
Their attacking setup leaves space behind, especially with aggressive full-backs.
Counter: Play through the lines quickly and use fast wingers to break into the space left by overlapping defenders.

Defensive Solidity:
Be prepared for a possession-based style with short passes.
Counter: Stay compact, deny short passing options in midfield, and force them wide where you can deal with crosses, rather than allowing central penetration.

Watch the Right Side:
Baldock and Basham provided aggressive overlaps on the right.
Counter: Double-team or contain these players to limit their impact and prevent them from getting into advanced areas.

In essence: Be direct, overload their aggressive flanks, press their build-up play, and stay defensively disciplined to counter their possession tactics.

So basically, AI thinks all you had to do to beat us in 2025 was to stop two men who last kicked a football in 2021 and 2023 and who had tragically passed away in 2024.

The frightening thing is not a single one of the responses on the thread spotted or questioned the obvious errors.

Is it any wonder our AI recruitment strategy didn't end well?
 
Following on from Cassius Kray's very sensible and timely reminder about the limitations and dangers of trusting AI for sourcing facts or for providing analysis (message #66 above), it reminded me of the tactical analysis I saw on the match thread of our recent home game against Birmingham City on one of their forums . . .


To save you having to soil yourselves on their grubby forum, here's what it said . . .



So basically, AI thinks all you had to do to beat us in 2025 was to stop two men who last kicked a football in 2021 and 2023 and who had tragically passed away in 2024.

The frightening thing is not a single one of the responses on the thread spotted or questioned the obvious errors.

Is it any wonder our AI recruitment strategy didn't end well?
Be why they got hammered 3-0 then.
 
Wily Boly is on a free in the Summer, might just be a case of taking over his contract until the end of the season, so not taking up a loan spot.

Interesting that Oxford United accidentally leaked their transfer targets on X over the weekend.
I think Matos and (surprisingly) Bindon were on there. There might be nothing in it, but maybe he's been told we might be sending him back.
& Rak-Sakyi on the Oxford targets list leak
 
Following on from Cassius Kray's very sensible and timely reminder about the limitations and dangers of trusting AI for sourcing facts or for providing analysis (message #66 above), it reminded me of the tactical analysis I saw on the match thread of our recent home game against Birmingham City on one of their forums . . .


To save you having to soil yourselves on their grubby forum, here's what it said . . .



So basically, AI thinks all you had to do to beat us in 2025 was to stop two men who last kicked a football in 2021 and 2023 and who had tragically passed away in 2024.

The frightening thing is not a single one of the responses on the thread spotted or questioned the obvious errors.

Is it any wonder our AI recruitment strategy didn't end well?
That’s generative AI in a nutshell.
Frightening how inaccurate it is and what people take for granted without sense checking it.

It’s good for summarising meetings and harvesting current company info but ask it to strategise and it lies and fabricates like a candidate on The Apprentice.
 



That’s generative AI in a nutshell.
Frightening how inaccurate it is and what people take for granted without sense checking it.

It’s good for summarising meetings and harvesting current company info but ask it to strategise and it lies and fabricates like a candidate on The Apprentice.
I used to work with a CEO of a group of companies (I say work with, I was basically a lackey) who rewrote absolutely everything with ChatGPT. Every email, document and presentation was put through and rewritten with the prompt "make this sound more formal/professional, whether it was to employees, lawyers or other CEOs. His profile on the website was written by generative AI. It hid his inability to write really well, and as far as I'm aware he's still doing it.

In an unrelated note, all 3 companies in the group went bust and he had to give away the last one for free.

If you know what to look for, it's also glaringly obvious when something is written (or even re-written) by generative AI.
 

All advertisments are hidden for logged in members, why not log in/register?

All advertisments are hidden for logged in members, why not log in/register?

Back
Top Bottom