Age Estimation is Challenging

(Part of the biometric product marketing expert series)

Two Biometric Update stories that were published on March 27, 2025 reminded me of something I wrote before.

One involved Paravision.

An announcement from Paravision says its biometric age estimation technology has achieved Level 3 certification from the Age Check Certification Scheme (ACCS), the leading independent certification body for age estimation. The results make it one of only six companies globally to receive ACCS’s highest-level designation for compliance.

San Francisco-based Paravision’s age estimation tech posted 100 percent precision in Challenge 25 compliance, with 0 subjects falsely identified as over 25 years old. It also scored a 0 percent Failure to Acquire Rate, meaning that every image submitted for analysis returned a result. Mean Absolute Error (MAE) was 1.37 years, with Standard Deviation of 1.17.

Now this is an impressive achievement, and Paravision is a quality company, and Joey Pritikin is a quality biometric executive, but…well, let me share the other story first, involving a Yoti customer (not Yoti).

Fenix responded that it set a challenge threshold at 23 years of age. Any user estimated to be that age or younger based on their face biometrics is required to use a secondary method for age verification.

Fenix had set OnlyFans challenge age, it turns out, at 20 years old. A correction to 23 years old was carried out on January 16, and then Fenix changed it again three days later, to 21 years old, Ofcom says.

Now Biometric Update was very clear that “Yoti provides the tech, but does not set the threshold.”

Challenge ages and legal ages

But do challenge thresholds have any meaning? I addressed that issue back in May 2024.

Many of the tests used a “Challenge-T” policy, such as “Challenge 25.” In other words, the test doesn’t estimate whether a person IS a particular age, but whether a person is WELL ABOVE a particular age….

So if you have to be 21 to access a good or service, the algorithm doesn’t estimate if you are over 21. Instead, it estimates whether you are over 25. If the algorithm thinks you’re over 25, you’re good to go. If it thinks you’re 24, pull out your ID card.

And if you want to be more accurate, raise the challenge age from 25 to 28.

NIST admits that this procedure results in a “tradeoff between protecting young people and inconveniencing older subjects” (where “older” is someone who is above the legal age but below the challenge age).

You may be asking why the algorithms have to set a challenge age above the lawful age, thus inconveniencing people above the lawful age but below the challenge age.

The reason is simple.

Age estimation is not all that accurate.

I mean, it’s accurate enough if I (a person well above the age of 21 years) must indicate whether I’m old enough to drink, but it’s not sufficiently accurate for a drinker on their 21st birthday (in the U.S.), or a 13 year old getting their first social media account (where lawful).

Not an official document.

If you have a government issued ID, age verification based upon that ID is a much better (albeit less convenient) solution.

(Kid computer picture by Adrian Pingstone – Transferred from en.wikipedia, Public Domain, https://commons.wikimedia.org/w/index.php?curid=112727.)

(Fake driver license picture from https://www.etsy.com/listing/1511398513/editable-little-drivers-license.)

Monitoring the #connectid hashtag

I have a long history with hashtags.

A LONG history.

Fires and parades

How long?

Back on October 23, 2007, I used my then-active Twitter account to tweet about the #sandiegofire. The San Diego fire was arguably the first mass adoption of hashtags, building upon pioneering work by Stowe Boyd and Chris Messina and acted upon by Nate Ritter and others.

From https://twitter.com/oemperor/status/358071562. Frozen peas? Long story.

The tinyurl link directed followers to my post detailing how the aforementioned San Diego Fire was displacing sports teams, including the San Diego Chargers. (Yes, kids, the Chargers used to play in San Diego.)

So while I was there at the beginning of hashtags, I’m proudest of the post that I wrote a couple of months later, entitled “Hashtagging Challenges When Events Occur at Different Times in Different Locations.” It describes the challenges of talking about the Rose Parade when someone is viewing the beginning of the parade while someone else is viewing the end of the parade at the same time. (This post was cited on PBWorks long ago, referenced deep in a Stowe Boyd post, and cited elsewhere.)

Hashtag use in business

Of course, hashtags have changed a lot since 2007-2008. After some resistance, Twitter formally supported the use of hashtags, and Facebook and other services followed, leading to mass adoption beyond the Factory Joes of the world.

Ignoring personal applications for the moment, hashtags have proven helpful for business purposes, especially when a particular event is taking place. No, not a fire in a major American city, but a conference of some sort. Conferences of all types have rushed to adopt hashtags so that conference attendees will promote their conference attendance. The general rule is that the more techie the conference, the more likely the attendees will use the conference-promoted hashtag.

I held various social media responsibilities during my years at MorphoTrak and IDEMIA, some of which were directly connected to the company’s annual user conference, and some of which were connected to the company’s attendance at other events. Obviously we pulled out the stops for our own conferences, including adopting hashtags that coincided with the conference theme.

A tweet https://twitter.com/JEBredCal/status/1124159756157849600 from the last (obviously celebratory) night of IDEMIA’s (Printrak’s) 40th conference in 2019. Coincidentally, this conference was held in San Diego.

And then when the conference organizers adopt a hashtag, they fervently hope that people will actually USE the adopted hashtag. As I said before, this isn’t an issue for the technical conferences, but it can be an issue at the semi-technical conferences. (“Hey, everybody! Gather around the screen! Someone used the conference hashtag…oh wait a minute, that’s my burner account.”)

A pleasant surprise with exhibitor/speaker adoption of the #connectID hashtag

Well, I think that we’ve finally crossed a threshold in the biometric world, and hashtags are becoming more and more acceptable.

As I previously mentioned, I’m not attending next week’s connect:ID conference in Washington DC, but I’m obviously interested in the proceedings.

So I turned to Twitter to check if anyone was using a #connectID hashtag in advance of the event. (Helpful hint: hashtags cannot include special characters such as “:” so don’t try to tweet #connect:ID; it won’t work and will appear as #connect.) Using the date-sorted search https://twitter.com/search?q=%23connectid&src=typed_query&f=live, I was expecting to see a couple of companies using the hashtag…if I was lucky.

I was pleasantly surprised to see that nearly two dozen exhibitors and speakers were using the #connectID hashtag (or referenced via the hashtag) as of the Friday before the event, including Acuity Market Intelligence, Aware, BIO-key, Blink Identity, Clearview AI, HID Global, IDEMIA, Integrated Biometrics, iProov, Iris ID, Kantara, NEC and NEC NSS, Pangiam, Paravision, The Paypers, WCC, WorldReach Software/Entrust, and probably some others by the time you read this, as well as some others that I may have missed.

And the event hasn’t even started yet.

At least some of the companies will have the presence of mind to tweet DURING the event on Tuesday and Wednesday.

Will yours be one of them?

But company adoption is only half the battle

While encouraging to me, adoption of a hashtag by a conference’s organizers, exhibitors, and speakers is only the beginning.

The true test will take place when (if) the ATTENDEES at the conference also choose to adopt the conference hashtag.

According to Terrapin (handling the logistics of conference organization), more than 2,500 people are registered for the conference. While the majority of these people are attending the free exhibition, over 750 of them are designated as “conference delegates” who will attend the speaking sessions.

How many of these people will tweet or post about #connectID?

We’ll all find out on Tuesday.

connect:ID 2021 is coming

I have not been to an identity trade show in years, and sadly I won’t be in Washington DC next week for connect:ID…although I’ll be thinking about it.

I’ve only been to connect:ID once, in 2015. Back in those days I was a strategic marketer with MorphoTrak, and we were demonstrating the MorphoWay. No, not the Morpho Way; the MorphoWay.

At connect:ID 2015.

Perhaps you’ve seen the video.

Video by Biometric Update. https://www.youtube.com/watch?v=mqfHAc227As

As an aside, you’ll notice how big MorphoWay is…which renders it impractical for use in U.S. airports, since space is valuable and therefore security features need a minimum footprint. MorphoWay has a maximum footprint…just ask the tradespeople who were responsible for getting it on and off the trade show floor.

I still remember several other things from this conference. For example, in those days one of Safran’s biometric competitors was 3M. Of course both Safran and 3M have exited the biometric industry, but at the time they were competing against each other. Companies always make a point of checking out the other companies at these conferences, but when I went to 3M’s booth, the one person I knew best (Teresa Wu) was not at the booth. Later that year, Teresa would leave 3M and (re)join Safran, where she remains to this day.

Yes, there is a lot of movement of people between firms. Looking over the companies in the connect:ID 2021 Exhibitor Directory, I know people at a number of these firms. Obviously people from IDEMIA, of course (IDEMIA was the company that bought Safran’s identity business), but I also know people at other companies, all of whom who were former coworkers at IDEMIA or one of its predecessor companies:

  • Aware.
  • Clearview AI.
  • GET Group North America.
  • HID Global.
  • Integrated Biometrics.
  • iProov.
  • NEC.
  • Paravision.
  • Rank One Computing.
  • SAFR/RealNetworks.
  • Thales.
  • Probably some others that I missed.

And I know people at some of the other companies, organizations, and governmental entities that are at connect:ID this year.

Some of these entities didn’t even exist when I was at connect:ID six years ago, and some of these entities (such as Thales) have entered the identity market due to acquisitions (in Thales’ case, the acquisition of Gemalto, which had acquired 3M’s biometric business).

So while I’m not crossing the country next week, I’m obviously thinking of everything that will be going on there.

Incidentally, this is one of the last events of the trade show season, which is starting to wind down for the year. But it will ramp up again next spring (for you Northern Hemisphere folks).

Bredemarket remembers the Southern Hemisphere, even though Bredemarket only does business in the United States. https://www.youtube.com/watch?v=HtZCQiN3n50

Regardless of where you are, hopefully the upcoming trade show season will not be adversely impacted by the pandemic.