The Blue Pill.

[BIAD]

Ignoring the diversity-quota that the BBC have to fill, here's an article that lightly shoehorn's in the future
of the UK's surveillance techniques.

It's not about gruff-sounding bank-robbers avoiding the long-arm of the law and you can be sure facial
recognition has no place in seeking the four or five well-dressed men leaving a Government-funded
apartment with smiles on their faces and a child's name on their minds.

Nope... it's about terrorism, the latest import to reach Great Britain's shores.
Instead of coming to terms with the knowledge that a virus-like violent culture is slowly consuming what
was once a proud and sympathetic social global community and administering a mature, lasting cure,
the BBC Blue-Haired Brigade are raising their hands and hopes in worship of a technology that is a mere
spin-off from their narcissistic perception of their safe-spaced reality.

.......................................

The cameras that know if you're happy -or a threat.

'Facial recognition tech is becoming more sophisticated, with some firms claiming it can even read our
emotions and detect suspicious behaviour. But what implications does this have for privacy and civil liberties?


Affectiva says its algorithms can detect hidden emotions in facial expressions.

Facial recognition tech has been around for decades, but it has been progressing in leaps and bounds in
recent years due to advances in computing vision and artificial intelligence (AI), tech experts say.

It is now being used to identify people at borders, unlock smart phones, spot criminals, and authenticate
banking transactions. But some tech firms are claiming it can also assess our emotional state...'

Some tech firms -you say? Companies that seek free advertising through an alleged respectable news
outlets, perhaps? Great Scott...! this is almost like an Elon-Musk ruse to get his electric car ideas into
the mainstream consciousness!

Be aware, the first image from the web page fails to identify who is in this picture. It's Rana el Kaliouby, the
CEO of Affectiva. The company currently holds two million images of people's faces. Their technology is used
in date-collecting, marketing, speech-analysis and measuring people’s responses during a political debate.

One wonders why Ms. el Kaliouby isn't identified in the picture?

But please, continue and disarm the reader's concerns by mentioning the collection of physical identity
had been going on for decades.

'...Since the 1970s, psychologists say they have been able to detect hidden emotions by studying the
"micro expressions" on someone's face in photographs and video.
Algorithms and high definition cameras can handle this process just as accurately and faster, tech firms
say.

"You're already seeing it used for commercial purposes," explains Oliver Philippou, an expert in video
surveillance at IHS Markit. "A supermarket might use it in the aisles, not to identify people, but to analyse
who came in in terms of age and gender as well as their basic mood. It can help with targeted marketing
and product placement."

Market research agency Kantar Millward Brown uses tech developed by US firm Affectiva to assess how
consumers react to TV adverts. Affectiva records video of people's faces -with their permission -then "codes"
their expressions frame by frame to assess their mood.

"We interview people but we get much more nuance by also looking at their expressions.
You can see exactly which part of an advert is working well and the emotional response triggered," says
Graham Page, managing director of offer and innovation at Kantar Millward Brown.

Millward Brown is a British multinational market research firm that partnered with Affectiva in 2011.
This article is basically a deconstructed brochure-piece for the umbrella-company Kantar without the
reader knowing it.

'...More controversially, a crop of start-ups are offering "emotion detection" for security purposes.
UK firm WeSee, for example, claims its AI tech can actually spot suspicious behaviour by reading facial cues
imperceptible to the untrained eye...'

Suspicious...?! I thought this article was about how facial recognition could help in the positive side of life.

A WeSee report:
'WeSee, the deep learning-based computer vision innovator that can detect suspicious behaviour,
has been selected to participate in Pitch@Palace 9.0, the initiative founded by The Duke of York in
2014 as a platform to amplify and accelerate the work of Entrepreneurs.'

Hmmm... I think we've wandered off the snowflake-caressing here. Let's get back on-point.

'...Emotions, such as doubt and anger, might be hidden under the surface in contrast to the language a person is using.
WeSee says it has been working with a "high profile" organisation in law enforcement to analyse people who are
being interviewed.


The iPhone X can be unlocked using facial recognition.

"Using only low-quality video footage, our technology has the ability to determine an individual's state of mind or
intent through their facial expressions, posture, gestures and movement," chief executive David Fulton tells the BBC.

"In future, video cameras on a tube station platform could use our tech to detect suspicious behaviour and alert
authorities to a potential terrorist threat...'

That darn terrorist threat, it's everywhere. Maybe the only way forward is to install this technology into devices
already in the public domain. For scientific purposes -of course.

'..."The same could be done with crowds at events like football matches or political rallies." But Mr Philippou is
sceptical about the accuracy of emotion detection.
"When it comes simply to identifying faces, there are still decent margins of error -the best firms claim they can
identify people with 90%-92% accuracy. "When you try assess emotions, too, the margin of error gets significantly
bigger."...'

Phew, for a minute there, Butch... I thought we were in trouble. Being scanned for private thoughts at a
political rally or a sporting event without permission is quite worrying. Luckily, the computer magic isn't
up it it's task.

But what a shame, a large BBC web article and it's ruined by the subject matter not being up to scratch.
A more cynical person might think they're currently halfway through on one of those pseudo-reality television
programmes that imply an obstacle that might thwart it's continuation after the commercials.

Oh-Lordy, I hope the writing becomes more upbeat.

'...That worries privacy campaigners who fear facial recognition tech could make wrong or biased judgements.
"While I can imagine that there are some genuinely useful use-cases, the privacy implications stemming from emotional
surveillance, facial recognition and facial profiling are unprecedented," says Frederike Kaltheuner of Privacy International.
Straightforward facial recognition is controversial enough....'

Taken from the Privacy International website:

In this section, you can read our report ‘Teach ’em to Phish: State Sponsors of Surveillance’,
as well as access a range of other resources about how powerful governments are financing,
training and equipping countries —including authoritarian regimes —with surveillance capabilities.

’Teach ‘em to Phish’ warns that rather than increasing security, state ‘security assistance’ programmes
are entrenching authoritarianism, further facilitating human rights abuses against people, and diverting
resources from long-term development programmes.
Privacy International:

Well it seems Governments are not playing well with those they're supposed to serve.
Using the latest surveillance technology, those in power are oppressing their people using tax-payers monies.
The BBC is Government funded, isn't it?

'...When revellers attended BBC Radio 1's Biggest Weekend in Swansea in May, many will have been unaware
that their faces were being scanned as part of a huge surveillance operation by South Wales Police...'

Yeah, but it's all for a good cause -it's implied. I'm sure that's what Mengele used to say.
But eh, BBC... don''t report the without-permission Police action in the negative. At the very least, convince
the reader it was a good cause with some evidence.

'...The force had deployed its Automated Facial Recognition (AFR) system, which uses CCTV-type cameras
and NEC software to identify "people of interest", comparing their faces to a database of custody images.



One man on an outstanding warrant was identified and arrested "within 10 minutes" of the tech being deployed
at the music festival, says Scott Lloyd, the AFR project leader for South Wales Police...'

Wooo-hoo! Go-Big Brother!

'...But human rights group Liberty points out that the tech has yielded large numbers of "false positive" matches
at other events, such as the Champions League final in Cardiff last year...'

What...?! Are you saying this AI technology thinks a type of people all look the same? That's racist!
Oh my word, we need a march through London and quick about it. The weather might change.

'...And in July, Cardiff resident Ed Bridges -represented by Liberty -began legal action against the force,
arguing that AFR violated people's privacy and lacked proper scrutiny, paving the way for a High Court battle...'

That case will be struck down if the Judge is a lower Masonic degree than the Chief Constable Matt Jukes of
South Wales Police. Of course it's violation of people's privacy, that's why they did it.
Sh*t, this isn't going well, it's supposed to be watered-down towards the end of these things.

'...But the technology is becoming more reliable, says Patrick Grother, head of biometric testing at the National
Institute of Standards & Technology, a US federal agency that conducts research into facial recognition.
He attributes the recent technological progress to the development of "convolutional neural networks" 
an advanced form of machine learning that enables a much greater degree of accuracy...'

Whooah, I'm dizzy. We jumped from reporting in the UK to quoting somebody from the US that spying
on people is getting better. If a guy-with-a-tie from the Congo says using voodoo poison to monitor the
public is viable, it doesn't mean you can. Different countries have different laws.

Sorry, I'm ruining the bullsh*t narrative.

"These algorithms allow computers to analyse images at different scales and angles," he says.
"You can identify faces much more accurately, even if they are partially obscured by sunglasses or scarves.
The error rate has come down ten-fold since 2014, although no algorithm is perfect."


Chinese police recently began using sunglasses fitted with a facial recognition system.

WeSee's Mr Fulton says his tech is simply a tool to help people assess existing video footage more intelligently.
He adds that WeSee can detect emotion in faces as effectively as a human can - "with around 60%-70%
accuracy"...'

BBC:

Partially-obscured faces and can detect emotion as effectively as a human... and an image of law enforcement
in a different country using this technology.
This future must bring a smile to your face.