6 coppers to take him down or 1 member of the public with his bike? 
 

 

 

56 minutes ago, Ani said:

I Just read the full story. 
 

The Police commissioned research on the accuracy of findings and according to one of these exercises it showed that of the false findings black people were significantly more likely to be identified. There was another research exercise at the same time that said although the % of black folk identified was higher it was not that relevant as it could be cause by so few false positives. 
 

Given the findings the police switched off the recognition till a new algorithm was released and this has got rid of the problem.  
 

The article above is slightly misleading as by saying as what does EXTREMELY mean ? There were about 180 people in the sample and 6 false positives were retuned . Just under 25% in the sample were black, but they gave 4 of the 6 positive negatives which is 66%. So 25% of your sample is giving 66% of false positives were returned. 
 

Seems like a bit of a storm in a teacup to me. Probably caused as we know all black people look the same 😉🙄
 

https://www.theregister.com/2026/03/20/uk_police_force_suspend_live_faical_recog_racial_bias/#:~:text=It found that at the,systematic effect%2C" it said.

 

Can you tell I have an A Level in Statistics 😂

So it’s was getting the right people but getting more black folk, the same as it was getting more men than women, but they paused it because they didn’t want bias……even though it was correctly identifying folk .

43 minutes ago, royal white said:

So it’s was getting the right people but getting more black folk, the same as it was getting more men than women, but they paused it because they didn’t want bias……even though it was correctly identifying folk .

It was making mistakes so they updated the software. The point is that it was falsely identifying more black folk. So not getting all the right people. 
 

There were 180 in the sample and it gave 6 false positives. Depends what error rate is deemed acceptable and if it was felt that error rate was likely to increase if they increased the sample size. But the fact that black folk were more than twice as likely to be picked out in error is not good. 
 

Remember this was a test using a test sample of 180, so once they go to a full watchlist how many false positives can they manage. 
 

If black folk are more likely to be incorrectly identified as are men. The issue for black men becomes more exaggerated. But if you run an exercise to check your system is working properly and it shows an issue that be corrected by a software fix you are going to implement it. The unknown is how long to gap between switching off and back on is. Given the overall impact of the software has been positive and Govt is rolling out more got to assume the gap was not massive. 
 

Edit - Just noticed it picked up ‘about half’ of people on the list so about 90. 6 false positives from 90 so a 6.66%?error rate. 

Edited March 20Mar 20 by Ani

On 17/03/2026 at 13:36, Not in Crawley said:

Thanks. Well we are at the moment 😬 Just lucky my eldest wasn't at Chem as she goes after work on a saturday but she's been skint. 

Other two are hoping for a day off even though they only closed the sixth form at Langton. Odd hearing Wes Streeting memtion the school in the HoP though!

Hope you're all still well. Hopefully you're now more vigilant to the symptoms as well.

Problem is for many folk that you think you've just got a nasty flu, then it gets very serious very fast.

It nearly killed me when I was 17. Luckily I got away with no lasting damage - at least I think so 😄

39 minutes ago, Ani said:

It was making mistakes so they updated the software. The point is that it was falsely identifying more black folk. So not getting all the right people. 
 

There were 180 in the sample and it gave 6 false positives. Depends what error rate is deemed acceptable and if it was felt that error rate was likely to increase if they increased the sample size. But the fact that black folk were more than twice as likely to be picked out in error is not good. 
 

Remember this was a test using a test sample of 180, so once they go to a full watchlist how many false positives can they manage. 
 

If black folk are more likely to be incorrectly identified as are men. The issue for black men becomes more exaggerated. But if you run an exercise to check your system is working properly and it shows an issue that be corrected by a software fix you are going to implement it. The unknown is how long to gap between switching off and back on is. Given the overall impact of the software has been positive and Govt is rolling out more got to assume the gap was not massive. 
 

Edit - Just noticed it picked up ‘about half’ of people on the list so about 90. 6 false positives from 90 so a 6.66%?error rate. 

Thinking about it, if the programme is giving more false positives for black people, it is falsely identifying them, rather than the right black people so as well as identifying the ‘wrong black people’ it must also be not identifying the ‘right black people’ so black people on the list are less likely to be identified. So that means black criminals are less likely to be arrested than their white counterparts. 

22 minutes ago, London Wanderer said:

Hope you're all still well. Hopefully you're now more vigilant to the symptoms as well.

Problem is for many folk that you think you've just got a nasty flu, then it gets very serious very fast.

It nearly killed me when I was 17. Luckily I got away with no lasting damage - at least I think so 😄

Well, East Kent is bonkers anyway, its beyond yokel at times.

Its been my fun this week - the local groups have gone into overdrive from its all to do with immigrants to fumigate every place. Best one was, we've got to block to trains to Sittingbourne 😁

There is a reason we have a reform council here.

2 hours ago, Ani said:

I Just read the full story. 
 

The Police commissioned research on the accuracy of findings and according to one of these exercises it showed that of the false findings black people were significantly more likely to be identified. There was another research exercise at the same time that said although the % of black folk identified was higher it was not that relevant as it could be cause by so few false positives. 
 

Given the findings the police switched off the recognition till a new algorithm was released and this has got rid of the problem.  
 

The article above is slightly misleading as by saying as what does EXTREMELY mean ? There were about 180 people in the sample and 6 false positives were retuned . Just under 25% in the sample were black, but they gave 4 of the 6 positive negatives which is 66%. So 25% of your sample is giving 66% of false positives were returned. 
 

Seems like a bit of a storm in a teacup to me. Probably caused as we know all black people look the same 😉🙄
 

https://www.theregister.com/2026/03/20/uk_police_force_suspend_live_faical_recog_racial_bias/#:~:text=It found that at the,systematic effect%2C" it said.

 

Can you tell I have an A Level in Statistics 😂

So it's the technology that is racist?

Just now, Zico said:

So it's the technology that is racist?

No

1 minute ago, Not in Crawley said:

No

(It was a joke)

10 minutes ago, Zico said:

(It was a joke)

Sorry, at work and its boring.

2 minutes ago, Not in Crawley said:

Sorry, at work and its boring.

Not enjoying the viking wrestling show?

25 minutes ago, frank_spencer said:

Not enjoying the viking wrestling show?

Its not for me. But thats not the point. the audience love it so Ive done my job.

39 minutes ago, Not in Crawley said:

Its not for me. But thats not the point. the audience love it so Ive done my job.

Seeing it promoted had me intrigued as I love me some wrestling. Is it still on in may? That's my next time in the big smoke.

49 minutes ago, frank_spencer said:

Seeing it promoted had me intrigued as I love me some wrestling. Is it still on in may? That's my next time in the big smoke.

Its on tour - Preston is your closest

9 hours ago, frank_spencer said:

Seeing it promoted had me intrigued as I love me some wrestling. Is it still on in may? That's my next time in the big smoke.

Woooo

Ragnar Lothbrok doing a Ric Flare strut 

https://www.theguardian.com/world/2026/mar/20/home-office-investigates-company-linked-religious-sect-immigration-visas

Nutjobs everywhere. 

1 hour ago, Tonge moor green jacket said:

https://www.theguardian.com/world/2026/mar/20/home-office-investigates-company-linked-religious-sect-immigration-visas

Nutjobs everywhere. 

Good old-fashioned loons

This is sadly becoming the norm. People from other countries battling it out on English streets carrying Rambo knives. 
 

 

4 hours ago, jmjhb said:

Good old-fashioned loons

With a modern twist of illegals.

What on earth possesses folk to fall for shit like this?

Can make the moon disappear ffs.

Kinell, it's like some twat saying he can make women's cup size bigger.

https://www.theboltonnews.co.uk/news/25954154.knifeman-stabbed-good-friend-queens-park-bolton/?cx_testId=3&cx_testVariant=cx_1&cx_artPos=0&cx_experienceId=EXB973903H0O&cx_experienceActionId=showRecommendationsATW0IF8SPZDP184&ref=prai#cxrecs_s

And another.

Closer to home too.

 

8 hours ago, royal white said:

This is sadly becoming the norm. People from other countries battling it out on English streets carrying Rambo knives. 
 

 

Throw the doors open. Let more in.

11 hours ago, jmjhb said:

Good old-fashioned loons without moons

 

More loons burning ambulances now, in Golders Green.

All the usual political platitudes of course, but fuck all will change.

Calm as fuck, three blokes walk up and set them on fire. 

7 hours ago, Tonge moor green jacket said:

 

All the usual political platitudes of course, but fuck all will change.

What do you want to happen to change things for the better?

Other than the urgent need to stop the arrival of unvetted illegal immigrants. I’m not sure what else you’re expecting politicians to do? 

This will be condemned by politicians & will now be left for the police to sort out. Hopefully those responsible locked up for a long time.

 Arson attacks have been carried by all sorts of loonies in recent times. We’ve seen Jews targeted, mosques targeted , asylum hotels targeted…. We’ve even had morons on social media encouraging it. Not sure how we go about stopping it, but only ever focusing on one demographic isn’t going to sort it. 

13 hours ago, Tonge moor green jacket said:

More loons burning ambulances now, in Golders Green.

All the usual political platitudes of course, but fuck all will change.

Calm as fuck, three blokes walk up and set them on fire. 

Palestine Action by any chance? Probably odds on favourites.

10 minutes ago, bolty58 said:

Palestine Action by any chance? Probably odds on favourites.

If it is they won’t get done for it got some cracking lawyers . Smash a van through the doors of a building in Bristol could of killed anyone on the other side then smash the place up with sledgehammer threaten the security guards all on cctv and get found not guilty