Author: azeeadmin

Daniel Kim and Jay Lee, the two founders of AuditBoard, a Los Angeles-based provider of a risk and compliance software service for large businesses, grew up middle school friends in Cerritos, Calif.

It was from their hometown Los Angeles exurb, that Kim and Lee first began plotting how they would turn their experience working for PriceWaterhouseCoopers and Ernst & Young (respectively) into the software business that just managed to rake in $40 million in financing led by one of venture capital’s most-respected firms, Battery Ventures.

Kim, who had moved on from the world of the big four audit firms to take positions as the head of global audit at companies as diverse as the chip component manufacturer, International Rectifier and the surf and sportswear-focused clothing company, Quiksilver, had complained to his childhood friend about how little had changed in the auditing world since the two men first started working in the industry.

For Kim, the frustration that systems for accounting for risk and compliance — requirements under the Sarbanes Oxley Act passed in 2002, were still little more than Excel spreadsheets tracking information across different business lines.

He thought there had to be a better way for companies to manage their audit and compliance processes. So with Lee’s help, he set out to build one. The two men touted the company’s service and its ability to create an out-of-the-box system of record for all internal audit, compliance and risk teams.

“It had been ten years since I had left audit. I couldn’t believe there wasn’t a software for compliance and risk,” Lee said. “Companies still manage Sarbanes-Oxley in Excel.”

There are other tools out there, IBM has OpenPages and ThomsonReuters developed a tool for audit and risk and compliance, but these software services pre-dated Sarbanes-Oxley, and were not made with a modern organization in mind, according to Lee and Kim.

The company counts major clients like TripAdvisor, Lululemon, HD Supply, Express Scripts and Spirit Airlines, among its roster of customers and will use the funding led by Battery to further expand its sales and marketing and product development efforts.

“We were impressed with AuditBoard’s product and its customer traction. With more CFOs now turning to dedicated, cloud-based software tools for various tasks, from ERP to tax compliance to procurement, we see a big opportunity for AuditBoard to continue to grow,” said Michael Brown, a general partner with Battery Ventures and the latest board member on AuditBoard’s board of directors. “We have invested before in similar companies that sell technology to CFOs — ranging from Avalara* and Intacct* to Outlooksoft* and Bonfire*– and we are excited to partner with Daniel, Jay and their team, who have already built a significant business in a short amount of time.”

AuditBoard raised a small seed round from friends and family, and followed that up with Donnelly Financial Solutions, a strategic investor who partnered with AuditBoard in 2017 to further develop its Securities and Exchange Commission reporting and Sarbanes-Oxley toolkit.

Now, AuditBoard joins a growing list of Los Angeles business-focused software companies that are beginning to scale dramatically in the city.

Long known for its advertising, marketing, and entertainment technology companies, large business-to-business software vendors are cropping up across the Los Angeles region. In addition to AuditBoard’s big round, companies like ServiceTitan, which raised $62 million in funding through an investment round led by Battery Ventures earlier in the year, are also making a splash in the Los Angeles business tech scene.

Earlier big rounds for companies like InAuth, the security firm; Factual, a location-based targeting service; PatientPop, the management tool for physicians offices; RightScale, a cloud management and cost optimization service; and Oblong Industries, a collaboration and computer interface developer, all speak to the breadth of the business-to-business talent that’s emerging from Hollywoodland.

 

Uber agreed to sell its Southeast Asia business in March, but it isn’t leaving the region. In fact, the U.S. firm is doubling down with plans to more than double its staff in Singapore.

That’s right. Uber is currently in the midst of a major recruitment drive that will see Singapore, the first city it expanded to in Asia, remain its headquarters for the Asia Pacific region despite its local exit. Unfortunately for customers who miss having a strong alternative to Grab, Uber won’t be bringing its ride-hailing app back in Singapore or anywhere else in Southeast Asia.

Uber’s own job portal lists 19 open roles for Singapore, but the company has contacted headhunting and recruitment firms to help fill as many as 75 vacancies, three sources with knowledge of Uber’s hiring plans told TechCrunch.

The new hires will take Uber’s headcount in Singapore to well over 100 employees, the sources claimed.

Ironically, of course, Uber let most of its staff in Southeast Asia leave when it stopped serving customers across its eight markets in Southeast Asia in April — although it was forced to extend into May in Singapore. As part of its exit deal, Grab got first dibs on 500 or so Uber Southeast Asia staff but that strategy didn’t pan out as planned, as TechCrunch previously reported. Indeed, a recent report suggested that fewer than 10 percent of ‘Uberites’ moved over to become ‘Grabbers’.

And yet, here we are, Uber is aggressively hiring in Singapore — but why?

The original plan following the Grab deal was for Uber to relocate its regional headquarters to either Japan or Hong Kong, two sources told TechCrunch, but in recent months that strategy has shifted. Just weeks ago, the remaining Singapore Uber collective — which consists of managers and executives — secured budget to staff up and find a larger office in the name of creating a support team for its remaining Asia Pacific markets.

The plan is for the Singapore-based employees to provide services such as HR, accounting, admin, marketing and PR across Uber APAC, which includes Hong Kong, Taiwan, Japan, Korea, Australia and India — although the latter has more sovereignty with its own president who reports into the U.S..

An Uber spokesperson acknowledged that the company is in the process of hiring in Singapore, but declined to provide further details.

Sources with knowledge of discussions inside the company told TechCrunch that the decision to stay in Singapore is down to a number of reasons.

Hong Kong, which had been a frontrunner to become Uber’s new APAC HQ, was ruled because Uber’s legal status in the country is unclear — a number of drivers have been prosecuted — while Japan and Australia were deemed to be too remote to be regional hubs. That left Singapore, as an established city for business with an existing Uber staff, as the remaining option.

Sources also told TechCrunch, however, that a degree of self-service was involved. Those executives and managers who managed to remove themselves from the “shame” of being shipped to Grab dug their heels in to avoid relocating their lives and families elsewhere, two sources claimed.

Talking to TechCrunch, some former Uber staff questioned whether the remaining Asian markets require remote services from Singapore, which is one of the world’s most expensive cities. Together the countries are hardly huge revenue generators for Uber and could be handled locally or other global cities. There’s certainly an argument that the continued investment in Singapore is at odds with the widely-held theory that Uber left Southeast Asia, a money-losing market, to clean up its balance sheet ahead of a much-anticipated IPO next year.

One former Uber employee who did transition to Grab noticed that the U.S. firm is now hiring for their previous role. That situation is made worse by a ban that prevented Uber’s Southeast Asia employees from applying to transfer to other parts of the firm’s global business. That’s despite many being allowed to do so in the case of previous Uber exit deals in China and Russia.

The result is that Uber is hiring in Singapore, a market where it no longer offers its service and gave up most of its staff to its rival. Anything can happen in the ride-sharing space!

Anyone with even a tangential relationship to the European tech startup scene knows that Startup Battlefield is one of the most effective launching pads for early-stage startups. All the pitch-competition drama and excitement goes down at Disrupt Berlin 2018 on November 29-30. If you want to spotlight your startup in front of the continent’s brightest innovators, investors and influencers, you have only one week left to submit your application — right here.

Last year at Disrupt Berlin 2017, Lia Diagnostics —  makers of the first flushable pregnancy test — won the Startup Battlefield and walked away with the Disrupt Cup, the $50,000 grand prize and an incredible amount of media coverage and investor interest. Could 2018 be your year?

Here’s what you need to know about competing in Startup Battlefield.

Our TechCrunch editors, steeped in the ways of identifying hot prospects since 2007, will review every application and select approximately 15 early-stage startups. Our acceptance rate typically hovers around three percent.

Participating founders receive free pitch coaching — again, from our Battlefield-tested editors — and they’ll be thoroughly prepped to step onto the TechCrunch Main Stage. That’s when the fun really starts. Teams have just six minutes to present a live demo to a distinguished panel of investors and entrepreneurs. Following each pitch, the judges have six minutes to grill the team with probing questions. That pitch coaching will come in handy, you betcha!

Only five teams move on to pitch a second time to a new panel of judges, and — after much discussion, conferring of notes and maybe an arm wrestle or two — the judges will choose one Startup Battlefield champion.

The entire competition takes place in front of a live audience — filled with thousands of people, including potential investors and customers. And plenty of media outlets, of course. Plus, we live-stream Startup Battlefield to a global audience (and make it available later, on-demand) on TechCrunch.com, YouTube, Facebook and Twitter.

That kind of exposure carries long-term benefits for all participating startups — not just the winner. It even has the potential to be life-changing.

And, because TechCrunch doesn’t charge any application or participation fees and we don’t take any equity from startups, you literally have nothing to lose by applying. Worse-case scenario: you don’t get to compete. Best-case scenario: your startup grows into the next unicorn. Hey, it could happen.

Startup Battlefield takes place at Disrupt Berlin 2018 on November 29-30. This is the perfect opportunity to introduce your startup to influencers across Europe and beyond. You have one week left before the application window closes. Apply to compete in Startup Battlefield today.

Karma, the Stockholm-based startup that offers a marketplace to let local restaurants and grocery offer unsold food at a discount, has raised $12 million in Series A funding.

Swedish investment firm Kinnevik led the round, with participation from U.S. venture capital firm Bessemer Venture Partners, appliance manufacturer Electrolux, and previous backer VC firm e.ventures. It brings total funding to $18 million.

Founded in late 2015 by Hjalmar Ståhlberg Nordegren, Ludvig Berling, Mattis Larsson and Elsa Bernadotte, and launched the following year, Karma is an app-based marketplace that helps restaurants and grocery stores reduce food waste by selling unsold food at a discount direct to consumers.

You simply register your location with the iOS or Android app and can browse various food merchants and the food items/dishes they have put on sale. Once you find an item to your liking, you pay through the Karma app and pick up the food before closing time. You can also follow your favourite establishments and be alerted when new food is listed each day.

“One third of of all food produced is wasted,” Karma CEO Ståhlberg Nordegren tells me. “We’re reducing food waste by enabling restaurants and grocery stores to sell their surplus food through our app… Consumers like you and me can then buy the food directly through the app and pick it up as take away at the location. We’re helping the seller reduce food waste and increase revenue, consumers get great food at a reduced price, and we help the environment redistributing food instead of wasting it”.

Since Karma’s original launch in its home country of Sweden, the startup has expanded to work with over 1,500 restaurants, grocery stores, hotels, cafes and bakeries to help reduce food waste by selling surplus food to 350,000 Karma users. It counts three of Sweden’s largest supermarkets as marketplace partners, as well as premium restaurants such as Ruta Baga and Marcus Samuelsson’s Kitchen & Table, and major brands such as Sodexo, Radisson and Scandic Hotels.

In February, the company expanded to the U.K., and is already working with over 400 restaurants in London. They include brands such as Aubaine, Polpo, Caravan, K10, Taylor St Barista’s, Ned’s Noodle Bar, and Detox Kitchen.

Ståhlberg Nordegren says Karma’s most frequent users are young professionals between the age of 25-40, who typically work in the city and pick up Karma on their way home. “Students and the elderly also love the app as it’s a great way to discover really good food for less,” he adds.

Meanwhile, will use the funding to continue to develop its product range, especially within supermarkets, and to expand to new markets, starting with Europe. The company plans to expand from 35 people based in Stockholm today to over 100 across 5 markets by the end of next year and over 150 by mid 2020.

Twitter has finally taken action against Infowars creator Alex Jones, but it isn’t what you might think.

While Apple, Facebook, Google/YouTube, Spotify and many others have removed Jones and his conspiracy-peddling organization Infowars from their platforms, Twitter has remained unmoved with its claim that Jones hasn’t violated rules on its platform.

That was helped in no small way by the mysterious removal of some tweets last week, but now Jones has been found to have violated Twitter’s rules, as CNET first noted.

Twitter is punishing Jones for a tweet that violates its community standards but it isn’t locking him out forever. Instead, a spokesperson for the company confirmed that Jones’ account is in “read-only mode” for up to seven days.

That means he will still be able to use the service and look up content via his account, but he’ll be unable to engage with it. That means no tweets, likes, retweets, comments, etc. He’s also been ordered to delete the offending tweet — more on that below — in order to qualify for a fully functioning account again.

That restoration doesn’t happen immediately, though. Twitter policy states that the read-only sin bin can last for up to seven days “depending on the nature of the violation.” We’re imagining Jones got the full one-week penalty, but we’re waiting on Twitter to confirm that.

The offending tweet in question is a link to a story claiming President “Trump must take action against web censorship.” It looks like the tweet has already been deleted, but not before Twitter judged that it violates its policy on abuse:

Abuse: You may not engage in the targeted harassment of someone, or incite other people to do so. We consider abusive behavior an attempt to harass, intimidate, or silence someone else’s voice.

When you consider the things Infowars and Jones have said or written — 9/11 conspiracies, harassment of Sandy Hook victim families and more — the content in question seems fairly innocuous. Indeed, you could look at President Trump’s tweets and find seemingly more punishable content without much difficulty.

But here we are.

The weirdest part of this Twitter caning is one of the reference points that the company gave to media. These days, it is common for the company to point reporters to specific tweets that it believes encapsulate its position on an issue, or provide additional color in certain situations.

In this case, Twitter pointed us — and presumably other reporters — to this tweet from Infowars’ Paul Joseph Watson:

WTF, Twitter…

U.S. accelerator Y Combinator is expanding to China after it announced the hiring of former Microsoft and Baidu Qi Lu who will develop a standalone startup program that runs on Chinese soil.

Shanghai-born Lu spent 11 years with Yahoo and eight years with Microsoft before a short spell with Baidu, where he was COO and head of the firm’s AI research division. Now he becomes founding CEO of YC China while he’s also stepping into the role of Head of YC Research. YC will also expand its research team with an office in Seattle, where Lu has plenty of links.

There’s no immediate timeframe for when YC will launch its China program, which represents its first global expansion, but YC President Sam Altman told TechCrunch in an interview that the program will be based in Beijing once it is up and running. Altman said Lu will use his network and YC’s growing presence in China — it ran its first ‘Startup School’ event in Beijing earlier this year — to recruit prospects who will be put into the upcoming winter program in the U.S..

Following that, YC will work to launch the China-based program as soon as possible. It appears that the details are still being sketched out, although Altman did confirm it will run independently but may lean on local partners for help. The YC President he envisages batch programming in the U.S. and China overlapping to a point with visitors, shared mentors and potentially other interaction between the two.

China’s startup scene has grown massively in recent years, numerous reports peg it close to that of the U.S., so it makes sense that YC, as an ‘ecosystem builder,’ wants to in. But Altman believes that the benefits extend beyond YC and will strengthen its network of founders, which spans more than 1,700 startups.

“The number one asset YC has is a very special founder community,” he told TechCrunch. “The opportunity to include a lot more Chinese founders seems super valuable to everyone. Over the next decade, a significant portion of the tech companies started will be from the U.S. or China [so operating a] network across both is a huge deal.”

Altman said he’s also banking on Lu being the man to make YC China happen. He revealed that he’s spent a decade trying to hire Lu, who he described as “one of the most impressive technologists I know.”

Entering China as a foreign entity is never easy, and in the venture world it is particularly tricky because China already has an advanced ecosystem of firms with their own networks for founders, particularly in the early-stage space. But Altman is confident that YC’s global reach and roster of founders and mentors appeals to startups in China.

YC has been working to add Chinese startups to its U.S.-based programs for some time. Altman has long been keen on an expansion to China, as he discussed at our Disrupt event last year, and partner Eric Migicovsky — who co-founder Pebble — has been busy developing networks and arranging events like the Beijing one to raise its profile.

That’s seen some progress with more teams from China — and other parts of the world — taking part in YC batches, which have never been more diverse. But YC is still missing out on global talent.

According to its own data, fewer than 10 Chinese companies have passed through its corridors but that list looks like it is missing some names so the number may be higher. Clearly, though, admission are skewed towards the U.S. — the question is whether Qi Lu and creation of YC China can significantly alter that.

Humans already find it unnerving enough when extremely alien-looking robots are kicked and interfered with, so one can only imagine how much worse it will be when they make unbroken eye contact and mirror your expressions while you heap abuse on them. This is the future we have selected.

The Simulative Emotional Experience Robot, or SEER, was on display at SIGGRAPH here in Vancouver, and it’s definitely an experience. The robot, a creation of Takayuki Todo, is a small humanoid head and neck that responds to the nearest person by making eye contact and imitating their expression.

It doesn’t sound like much, but it’s pretty complex to execute well, which despite a few glitches SEER managed to do.

At present it alternates between two modes: imitative and eye contact. Both, of course, rely on a nearby (or, one can imagine, built-in) camera that recognizes and tracks the features of your face in real time.

In imitative mode the positions of the viewer’s eyebrows and eyelids, and the position of their head, are mirrored by SEER. It’s not perfect — it occasionally freaks out or vibrates because of noisy face data — but when it worked it managed rather a good version of what I was giving it. Real humans are more expressive, naturally, but this little face with its creepily realistic eyes plunged deeply into the uncanny valley and nearly climbed the far side.

Eye contact mode has the robot moving on its own while, as you might guess, making uninterrupted eye contact with whoever is nearest. It’s a bit creepy, but not in the way that some robots are — when you’re looked at by inadequately modeled faces, it just feels like bad VFX. In this case it was more the surprising amount of empathy you suddenly feel for this little machine.

That’s largely due to the delicate, childlike, neutral sculpting of the face and highly realistic eyes. If an Amazon Echo had those eyes, you’d never forget it was listening to everything you say. You might even tell it your problems.

This is just an art project for now, but the tech behind it is definitely the kind of thing you can expect to be integrated with virtual assistants and the like in the near future. Whether that’s a good thing or a bad one I guess we’ll find out together.

While Elon Musk and Mark Zuckerberg debate the dangers of artificial general intelligence, startups applying AI to more narrowly defined problems such as accelerating the performance of sales teams and improving the operating efficiency of manufacturing lines are building billion-dollar businesses. Narrowly defining a problem, however, is only the first step to finding valuable business applications of AI.

To find the right opportunity around which to build an AI business, startups must apply the “Goldilocks principle” in several different dimensions to find the sweet spot that is “just right” to begin — not too far in one dimension, not too far in another. Here are some ways for aspiring startup founders to thread the needle with their AI strategy, based on what we’ve learned from working with thousands of AI startups.

 “Just right” prediction time horizons

Unlike pre-intelligence software, AI responds to the environment in which they operate; algorithms take in data and return an answer or prediction. Depending on the application, that prediction may describe an outcome in the near term, such as tomorrow’s weather, or an outcome many years in the future, such as whether a patient will develop cancer in 20 years. The time horizon of the algorithm’s prediction is critical to its usefulness and to whether it offers an opportunity to build defensibility.

Algorithms making predictions with long time horizons are difficult to evaluate and improve. For example, an algorithm may use the schedule of a contractor’s previous projects to predict that a particular construction project will fall six months behind schedule and go over budget by 20 percent. Until this new project is completed, the algorithm designer and end user can only tell whether the prediction is directionally correct — that is, whether the project is falling behind or costs are higher.

Even when the final project numbers end up very close to the predicted numbers, it will be difficult to complete the feedback loop and positively reinforce the algorithm. Many factors may influence complex systems like a construction project, making it difficult to A/B test the prediction to tease out the input variables from unknown confounding factors. The more complex the system, the longer it may take the algorithm to complete a reinforcement cycle, and the more difficult it becomes to precisely train the algorithm.

While many enterprise customers are open to piloting AI solutions, startups must be able to validate the algorithm’s performance in order to complete the sale. The most convincing way to validate an algorithm is by using the customer’s real-time data, but this approach may be difficult to achieve during a pilot. If the startup does get access to the customer’s data, the prediction time horizon should be short enough that the algorithm can be validated during the pilot period.

Historic data, if it’s available, can serve as a stopgap to train an algorithm and temporarily validate it via backtesting. Training an algorithm making long time horizon predictions on historic data is risky because processes and environments are more likely to have changed the further back you dig into historic records, making historic data sets less descriptive of present-day conditions.

In other cases, while the historic data describing outcomes exists for you to train an algorithm, it may not capture the input variable under consideration. In the construction example, that could mean that you found out that sites using blue safety hats are more likely to complete projects on time, but since that hat color wasn’t previously helpful in managing projects, that information wasn’t recorded in the archival records. This data must be captured from scratch, which further delays your time to market.

Instead of making singular “hero” predictions with long time horizons, AI startups should build multiple algorithms making smaller, simpler predictions with short time horizons. Decomposing an environment into simpler subsystems or processes limits the number of inputs, making them easier to control for confounding factors. The BIM 360 Project IQ Team at Autodesk takes this small prediction approach to areas that contribute to construction project delays. Their models predict safety and score vendor and subcontractor quality/reliability, all of which can be measured while a project is ongoing.

Shorter time horizons make it easier for the algorithm engineer to monitor its change in performance and take action to quickly improve it, instead of being limited to backtesting on historic data. The shorter the time horizon, the shorter the algorithm’s feedback loop will be. As each cycle through the feedback incrementally compounds the algorithm’s performance, shorter feedback loops are better for building defensibility. 

“Just right” actionability window

Most algorithms model dynamic systems and return a prediction for a human to act on. Depending on how quickly the system is changing, the algorithm’s output may not remain valid for very long: the prediction may “decay” before the user can take action. In order to be useful to the end user, the algorithm must be designed to accommodate the limitations of computing and human speed. 

In a typical AI-human workflow, the human feeds input data into the algorithm, the algorithm runs calculations on that input data and returns an output that predicts a certain outcome or recommends a course of action; the human interprets that information to decide on a course of action, then takes action. The time it takes the algorithm to compute an answer and the time it takes for a human to act on the output are the two largest bottlenecks in this workflow. 

For most of AI history, slow computational speeds have severely limited the scope of applied AI. An algorithm’s prediction depends on the input data, and the input data represents a snapshot in time at the moment it was recorded. If the environment described by the data changes faster than the algorithm can compute the input data, by the time the algorithm completes its computations and returns a prediction, the prediction will only describe a moment in the past and will not be actionable. For example, the algorithm behind the music app Shazam may have needed several hours to identify a song after first “hearing” it using the computational power of a Windows 95 computer. 

The rise of cloud computing and the development of hardware specially optimized for AI computations has dramatically broadened the scope of areas where applied AI is actionable and affordable. While macro tech advancements can greatly advance applied AI, the algorithm is not totally held hostage to current limits of computation; reinforcement through training also can improve the algorithm’s response time. The more of the same example an algorithm encounters, the more quickly it can skip computations to arrive at a prediction. Thanks to advances in computation and reinforcement, today Shazam takes less than 15 seconds to identify a song. 

Automating the decision and action also could help users make use of predictions that decay too quickly to wait for humans to respond. Opsani is one such company using AI to make decisions that are too numerous and fast-moving for humans to make effectively. Unlike human DevOps, who can only move so fast to optimize performance based on recommendations from an algorithm, Opsani applies AI to both identify and automatically improve operations of applications and cloud infrastructure so its customers can enjoy dramatically better performance.

Not all applications of AI can be completely automated, however, if the perceived risk is too high for end users to accept, or if regulations mandate that humans must approve the decision. 

“Just right” performance minimums

Just like software startups launch when they have built a minimum viable product (MVP) in order to collect actionable feedback from initial customers, AI startups should launch when they reach the minimum algorithmic performance (MAP) required by early adopters, so that the algorithm can be trained on more diverse and fresh data sets and avoid becoming overfit to a training set.

Most applications don’t require 100 percent accuracy to be valuable. For example, a fraud detection algorithm may only immediately catch five percent of fraud cases within 24 hours of when they occur, but human fraud investigators catch 15 percent of fraud cases after a month of analysis. In this case, the MAP is zero, because the fraud detection algorithm could serve as a first filter in order to reduce the number of cases the human investigators must process. The startup can go to market immediately in order to secure access to the large volume of fraud data used for training their algorithm. Over time, the algorithms’ accuracy will improve and reduce the burden on human investigators, freeing them to focus on the most complex cases.

Startups building algorithms for zero or low MAP applications will be able to launch quickly, but may be continuously looking over their shoulder for copycats, if these copycats appear before the algorithm has reached a high level of performance. 

Startups attacking low MAP problems also should watch out for problems that can be solved with near 100 percent accuracy with a very small training set, where the problem being modeled is relatively simple, with few dimensions to track and few possible variations in outcome.

AI-powered contract processing is a good example of an application where the algorithm’s performance plateaus quickly. There are thousands of contract types, but most of them share key fields: the parties involved, the items of value being exchanged, time frame, etc. Specific document types like mortgage applications or rental agreements are highly standardized in order to comply with regulation. Across multiple startups, we have seen algorithms that automatically process these documents needing only a few hundred examples to train to an acceptable degree of accuracy before additional examples do little to improve the algorithm, making it easy for new entrants to match incumbents and earlier entrants in performance.

AIs built for applications where human labor is inexpensive and able to easily achieve high accuracy may need to reach a higher MAP before they can find an early adopter. Tasks requiring fine motor skills, for example, have yet to be taken over by robots because human performance sets a very high MAP to overcome. When picking up an object, the AIs powering the robotic hand must gauge an object’s stiffness and weight with a high degree of accuracy, otherwise the hand will damage the object being handled. Humans can very accurately gauge these dimensions with almost no training. Startups attacking high MAP problems must invest more time and capital into acquiring enough data to reach MAP and launch. 

Threading the needle

Narrow AI can demonstrate impressive gains in a wide range of applications — in the research lab. Building a business around a narrow AI application, on the other hand, requires a new playbook. This process is heavily dependent on the specific use case on all dimensions, and the performance of the algorithm is merely one starting point. There’s no one-size-fits-all approach to moving an algorithm from the research lab to the market, but we hope these ideas will provide a useful blueprint for you to begin.

CEO John Lemp recently said that thanks to a new policy, publishers in Revcontent‘s content recommendation network “won’t ever make a cent” on false and misleading stories — at least, not from the network.

To achieve this, the company is relying on fact-checking provided by the Poynter Institute’s International Fact Checking Network. If any two independent fact checkers from International Fact Checking flag a story from the Revcontent network as false, the company’s widget will be removed, and Revcontent will not pay out any money on that story (not even revenue earned before the story was flagged).

In some ways, Revcontent’s approach to fighting fake news and misinformation sounds similar to the big social media companies — Lemp, like Twitter, has said his company cannot be the “arbiter of truth,” and like Facebook, he’s emphasizing the need to remove the financial incentives for posting sensationalistic-but-misleading stories.

However, Lemp (who’s spoken in the past about using content recommendations to reduce publishers’ reliance on individual platforms) criticized the big internet companies for “arbitrarily” taking down content in response to “bad PR.” In contrast, he said Revcontent will have a fully transparent approach, one that removes the financial rewards for fake news without silencing anyone.

Lemp didn’t mention any specific takedowns, but the big story these days is Infowars. It seems like nearly everyone has been cracking down on Alex Jones’ far-right, conspiracy-mongering site, removing at least some Infowars-related accounts and content in the past couple of weeks.

The Infowars story also raises the question of whether you can effectively fight fake news on a story-by-story basis, rather than completely cutting off publishers when they’ve shown themselves to consistently post misleading or falsified stories.

When asked about this, Lemp said Revcontent also has the option to completely removing publishers from the network, but he said he views that as a “last resort.”

The BitFi crypto wallet was supposed to be unhackable and none other than famous weirdo John McAfee claimed that the device – essentially an Android-based mini tablet – would withstand any attack. Spoiler alert: it couldn’t.

First, a bit of background. The $120 device launched at the beginning of this month to much fanfare. It consisted of a device that McAfee claimed contained no software or storage and was instead a standalone wallet similar to the Trezor. The website featured a bold claim by McAfee himself, one that would give a normal security researcher pause:

Further, the company offered a bug bounty that seems to be slowly being eroded by outside forces. They asked hackers to pull coins off of a specially prepared $10 wallet, a move that is uncommon in the world of bug bounties. They wrote:

We deposit coins into a Bitfi wallet
If you wish to participate in the bounty program, you will purchase a Bitfi wallet that is preloaded with coins for just an additional $10 (the reason for the charge is because we need to ensure serious inquiries only)
If you successfully extract the coins and empty the wallet, this would be considered a successful hack
You can then keep the coins and Bitfi will make a payment to you of $250,000
Please note that we grant anyone who participates in this bounty permission to use all possible attack vectors, including our servers, nodes, and our infrastructure

Hackers began attacking the device immediately, eventually hacking it to find the passphrase used to move crypto in and out of the the wallet. In a detailed set of tweets, security researchers Andrew Tierney and Alan Woodward began finding holes by attacking the operating system itself. However, this did not match the bounty to the letter, claimed BitFi, even though they did not actually ship any bounty-ready devices.

Then, to add insult to injury, the company earned a Pwnies award at security conference Defcon. The award was given for worst vendor response. As hackers began dismantling the device, BitFi went on the defensive, consistently claiming that their device was secure. And the hackers had a field day. One hacker, 15-year-old Saleem Rashid, was able to play Doom on the device.

The hacks kept coming. McAfee, for his part, kept refusing to accept the hacks as genuine.

Unfortunately, the latest hack may have just fulfilled all of BitFi’s requirements. Rashid and Tierney have been able to pull cash out of the wallet by hacking the passphrase, a primary requirement for the bounty. “We have sent the seed and phrase from the device to another server, it just gets sent using netcat, nothing fancy.” Tierney said. “We believe all conditions have been met.”

The end state of this crypto mess? BitFi did what most hacked crypto companies do: double down on the threats. In a recently deleted Tweet they made it clear that they were not to be messed with:

The researchers, however, may still have the last laugh.