Tag Archives: Bots

Stormtrooper bot with facial recognition guards against intruders

We know what Star Wars fans will be wishing for this Christmas — aside from Sphero’s R2-D2 and BB-9E toys. Ubtech, the company behind the dancing Lynx robot, is releasing a Stormtrooper bot. Although it can’t teach you yoga, you can use voice commands to instruct it to ward off intruders (hopefully not with an actual blaster). Its facial recognition tech also allows it to store up to three faces in its memory bank.

Image result for Stormtrooper

That way, it can shout at any uninvited guests that keep barging into your room. And, its accompanying mobile app packs an augmented reality game that lets you fight off the Resistance and launch First Order attacks. The robot is available for pre-order from Ubtech right now, and will ship in November. It will set you back $300.


Those keen to snap up the Stormtrooper should be aware that its manufacturer has previously come under fire from the security community. In August, cybersecurity firm IOActive claimed that Ubtech’s Alpha series models did not encrypt the data they stored, making it easy for hackers to steal the info. Tinkerers even managed to turn the cute little bots into screwdriver-wielding, stab-happy maniacs. Not exactly stocking filler material, then. But, no such claims have been made about the Star Wars First Order Stormtrooper.



Amazon bots are helping bring the USPS back from oblivion

The United States Postal Service is not exactly bouncing back from oblivion these days, but it’s worth noting that there’s been a slight bump in one area.

For the low-margin Shipping and Packages division, there was an 11 percent increase in revenue over last quarter. In general, these items — small packages that come to your mailbox but not snail mail letters — have kept the USPS from sinking like a ship.

For the savvy readers out there, you might already know there is one company that is helping them rebound. Hint: It’s named after an area in South America.

As Tim O’Reilly wisely notes in a recent post, Amazon is really pumping life into the shipping and receiving industry. Bots suggest products when we shop, they improve fulfillment, and they could one day guide drones to your doorstep. These are technically “automations” and not AI, but to a consumer, that doesn’t matter as long as that beef jerky and USB-C cable arrives faster, at a decent price, and in one piece.

O’Reilly also notes that this is saving jobs, not replacing them. I remember talking to an Amazon Now delivery person once a few months ago. In my area, I was able to order a printer cable and have it arrive in about an hour. He told me he was out of work for a few years. He was happy to be working, and he whistled as he walked out the door. Does he care that bots are making his job easier? Sure. Does he think bots will replace his job? Probably not. If anything, he’s happy they exist.

Next time you hear someone says bots are causing problems, tell them about that USPS increase since last quarter. You can blame automation for a lot of things, but at least in this case, it might be the bots that are saving the postal service.



A free robot lawyer is fighting parking tickets and much more

Lawyers can be really expensive, and for small disputes like fighting your landlord, claiming lost luggage for an airline, or a parking ticket, it can feel like a fight isn’t worth it.

Enter DoNotPay, the world’s first robot lawyer, built by young British entrepreneur Joshua Browder. DoNotPay hit headlines in early 2016 after then successfully appealing £2 million in parking tickets. The bot then expanded to help refugees, and now it’s expanding into 1,000 different areas of law in all 50 US states and across the UK.

According to MarketLine research, the US legal market alone is worth $292 billion (£227 billion).

Now Browder’s bot can help you ask for more parental leave, dispute nuisance calls, fight a fraudulent purchase on your credit card, and a host of other issues.

“I originally started DoNotPay two years ago to fight my own parking tickets and became an accidental witness to how lawyers are exploiting human misery,” said Browder. “From discrimination in Silicon Valley to the tragedy in London with an apartment building setting on fire, it seems the only people benefitting from injustice are a handful of lawyers.


“I hope that DoNotPay, by helping with these issues and many more, will ultimately give everyone the same legal power as the richest in society.”

The updated version of DoNotPay goes live on Wednesday.

Ask the bot a question and it generates a legal response for you

DoNotPay works as a Facebook chatbot, or through the DoNotPay site. You type in a query, like “I keep getting unwanted calls.” DoNotPay gives you a few options: it can generate a cease-and-desist letter for you to send to the relevant authorities, or if you’ve done that already, it can advise you on next steps.

Browder told Business Insider that the bot can’t take on court battles “yet”, but it can help with anything involving documents. It can fill PDFs, or generate letters, something he said is “easy to automate”. The bot is powered by IBM’s “Watson” AI technology.

That’s not to say training DoNotPay has been easy, particularly around natural language.

Browder said: “For example, knowing that when someone says ‘the signs were difficult to understand’ to the bot, it has to know to fill in the letter referring to incorrect signage regulations. This challenge was the hardest, but with Watson and a year of hard work, I am ecstatic that we made it possible.”

DoNotPay has also challenged considerably more parking tickets — it’s contested 375,000 tickets in the UK, New York, and Seattle, the equivalent of $10 million (£7.7 million).

Browder thinks DoNotPay might help prevent another tragedy like Grenfell, where at least 80 people died or went missing in a tower block fire.

“The current set of bots are designed to prevent further atrocities by making sure landlords and developers follow basic safety regulations,” he explained. “For example, to immediately fix a gas leak.”

The bot could also allow those left homeless by such accidents to claim housing more easily.

Browder’s bot has earned him national attention and, he said, it just started as a side project. He’s now entrepreneur-in-residence at venture firm Greylock, something he says he “could never have imagined.”



Google’s new reCAPTCHA automatically tells you’re not a bot

Over the years, Google has utilised a number of methods to distinguish between human and bots on the web. Its take on the CAPTCHA (Completely Automated Public Turing test to tell Computers and Humans Apart) test, known as reCAPTCHA, has required you to transcribe distorted words, confirm Street View addresses or simply just tick a box. Soon, you won’t need to do the hard work, because Google’s making the system invisible.

Using a combination of machine learning and advanced risk analysis, Google has updated its system to detect user habits without dedicated interaction. When you arrive on a web page, the controls should disappear and serve the relevant content. However, if you do trip Google’s risk analysis algorithms, you may need to quickly solve one of the search giant’s puzzles.

While the new system is invisible, it will still consider variables like your IP address and the movements of your mouse. Google says its technology will “actively consider a user’s engagement with the CAPTCHA — before, during, and after — to determine whether that user is a human.” That means no more transcription, which offered a human balance to Google’s optical character recognition, but you may now find what you were looking for a lot quicker.




Calgary author booted off Amazon, Internet bot the culprit

When Calgary author Adam Dreece chose to list his books exclusively with the world’s largest bookstore, he thought it would be a big boost to his sales.

Instead, he was taken down by an unexpected foe: an Internet bot.

Dreece, a bestselling author of books for young adults, said that in December he put his steampunk book series The Yellow Hoods up on Amazon Kindle Direct Publishing (KDP), a program requiring books enrolled be exclusive to Amazon.

Paying about $1,000 for online promos, Dreece said his books were steadily gaining traction. But after a sudden 25,000 page-read bump one day, Dreece said something didn’t feel right.

“I thought, this is fishy,” Dreece said. “I emailed Amazon and they came back to me and said everything was normal.”

But a few days later, Dreece received an email from the book retailer saying his account was being terminated and all his books were being removed from the online store due to downloads “originating from systematically generated accounts.”

Dreece’s entire Yellow Hoods series, book reviews, followers — everything was gone overnight.

Sean Gallagher, information technology editor with Ars Technica, said while Internet bots are nothing new, they can create a lot of problems for individuals such as Dreece trying to earn a living online.

“Amazon does a lot to try to screen fake accounts out, but they still happen,” Gallagher said. “If you’re selling through a marketplace that has a rating system like Amazon, bots could conceivably artificially drive your ratings (and your sales) down, or they could otherwise drive away business.”

There are a number of threads online with authors relaying stories of Amazon terminating their accounts for the same reason. Dreece said he had no idea why he was targeted or where these systematically generated accounts were coming from.

When asked about their methods for detecting fraud, Amazon said they try to be vigilant to protect authors.

“We regularly monitor for abuse across our programs. When we detect abuse (either direct or through third parties), we may terminate the account involved,” the company said in a statement. “If authors have questions or disagree with our findings, they contact us via the link provided at the end of the email they receive.”

Dreece said he was shocked when he received a call from Amazon a few weeks later saying they had made a mistake, and offered to reinstate his account.

But Dreece said the damage had already been done.

“I consider myself lucky. They restored my accounts, but they crushed my traffic from $20 per day to 18 cents,” he said.

While Gallagher said many bots are benign, there are some — such as the Mirai botnet scandal that infected digital cameras and subsequently took down Amazon, Twitter and Netflix last fall — that can cause real harm.

“There are bots used for all sorts of malicious activities, often spread through malware-infected PCs,” Gallagher said. “On Amazon, activity like this can be used to either artificially inflate the rating of an e-book or some other product to get it more notice, or to drive a title’s ratings down to knock it far down the listings, or get it withdrawn completely.”

Still confounded by the sudden, ill-fated attack, Dreece has decided not to put his books back on Amazon and will market himself elsewhere.

“Every book, everything I’ve built up over years, is gone,” Dreece said. “I think this is really the beginning of an era where scammers and bots are weapons now.”



Forget ‘user-friendly’ bots and focus on advanced usefulness

Just a few months into 2017 and we’ve already reached an inflection point for consumer automation software. From Alexa stealing the show at CES to an explosion in bot development from major companies, the impact of bots can be widely felt.


The path for the bot revolution has been laid by four converging trends. It started with app fatigue, as consumers recognized that while there may be an “app for everything,” there are really only three or four we use on a daily basis. Then one form of application that picked up enormous traction is messaging, so every utility imaginable has been adding messaging components just to get a share of people’s attention.


That increased interest in development around messaging has resulted in new technology frameworks from Microsoft, Facebook, Slack, and IBM — all focused on making smart software agent development nearly as simple as building a WordPress site. Lastly, we’ve become more comfortable talking with bots. While Spike Jonze’s Her bot may be a ways off, Siri, Alexa, and Google Assistant have all been carrying on useful conversations with consumers, often to their great delight.

The market for artificial intelligence — of which bots are a major subset  —  is expected to become a multibillion-dollar industry over the next decade. VentureBeat published an own overview of the market and discovered over 170+ companies, $4 billion in funding, and thousands of bots currently available.

Giants like Facebook and Slack are opening their platforms up to bots; startups are flooding the market with new skills and bots; and even the old guard, like Microsoft and IBM, are looking to get in on the action. As with all technology, the mass emergence of bots will take time. But the advances in voice technology have been greater in the last 18 months than in the last 30. And in the tech world, that’s a serious trend.

But if there is so much potential, why isn’t it being unleashed?

Developing the bot itself is just the beginning of an intelligent system. And building those are hard.

You need:

  1. Centralized, interfaceable data — Not just from one supplier, but from a range of data holders all working together. Now think how hard companies have been working to centralize profile data in one place, combine sales and marketing data, and then combine that with information about their customers outside their own domain.
  2. The smarts to predict behavior — Now that is real magic.  We are getting closer to predictive ambient systems that help us do things before we know we want to, like Netflix promoting shows it thinks I like and LinkedIn showing me people I may want to schmooze with. And we are getting to banks that advise me on how to spend my earnings and grocery stores that recommend product.
  3. A well-designed user experience — This is the hardest yet. Designing calm systems that are helpful, utilitarian, and yet engaging is the dream of UX.

Over the last few years, we in the bot community have all been focused on designing user-friendly software, but if we continue to design these bots to just be “friendly” we are missing a million opportunities. From personal trainers to bankers to doctors to insurance brokers, bots can help, but we need these agents to deliver more than friendliness. We need motivation, empathy, discretion, and relevance, at least.

Designing systems to interact with us at the right time in a calm fashion is going to get really hard really quick. Ultimately, we are designing bots to act like humans or to seem like they are humans. This is probably the toughest frontier for us as designers — machines need to be more than fast and accurate. They need to be useful in a way that’s even hard to quantify.

If we ever get there, how do we redesign our businesses to accommodate these new bots? Do we protect them like we do our employees when customers are rude to them? Do we make laws to govern them if the information they provide is wrong? And what do we do with bots when they are not useful anymore?

We need to design for a complex, messy world with competing demands, a world where user experience design means making something completely new.  And, as designers, we need to get really smart about understanding ourselves and our needs as we embark in this new world.

Toby Barnes is the Group Strategy Director at AKQA, an ideas and innovation company.



%d bloggers like this: