AI can determine if you land a job, get a loan, and now, remove gender bias from Disney movies.

Businesses across almost every industry deploy artificial intelligence to make jobs simpler for staff and tasks easier for consumers. For example, computer software teaches customer service agents how to be more compassionate, schools use machine learning to scan for weapons and mass shooters on campus, and doctors use AI to map the root cause of diseases. Though these applications may seem harmless, AI is only as good as the data it is fed.
Related image
One of the greatest obstacles facing the proliferation and acceptance of AI, is bias. As AI systems spread further and have influence over our lives, it’s getting more important to make sure they’re properly trained.  Whether it’s intentional or not, humans make judgments that can spill over into the code created for AI to follow. That means AI can contain implicit racial, gender and ideological biases, which in some cases can lead to flawed results and dangerous conclusions.
In its more controversial applications, AI is now being used to predict how likely a person is to commit a crime, how a person might behave on the job, and whether they’re worthy of borrowing money from a bank. Though these applications are helpful, they can disadvantage underrepresented groups if deployed incorrectly. While bias can creep-in easily in AI, it can also help to reduce disparities caused by poor human judgement.
Image result for ai bias
When you break down classic Disney movies, they usually have very problematic undertones. But in recent years Disney has become more progressive, moving away from the largely “Prince Charming”-type story lines. Now Disney has taken it a step further to improve its representation by turning to AI, using “GD-IQ” — a tool that reviews scripts to spellcheck gender bias. The tool, is being used to evaluate how many characters are part of the LGBTQ+ community, how many characters are people of color, how many have disabilities, as well as characters that are part of other minority groups that aren’t frequently represented in film and television.
Related image
Why it’s hot: AI is no longer the stuff of science fiction, and as such we need to become more conscious of AI’s underlying biases. Organizations can make tweaks to their algorithms and data sources, but at the end of the day, the final output is ultimately judged by a human. Only by adding more women, people of color and other underrepresented groups to the team to help implement these technologies can we create more equitable systems that address AI bias.

An Old Brand Learns a New Trick Ahead of Prime Day

Amazon is estimated to have pulled in $6-7 billion this year from its 5th annual “Prime Day”, but the e-commerce giant wasn’t the only brand to score big from one of the biggest spending occasions of the year.

Image result for prime day

Amazon’s recommendation algorithm has made it easy to discover and find products you never knew you needed; however, an amazing experience for the user doesn’t always translate to positive outcomes for competing brands. When you buy a product on Amazon, the brand you choose will be the brand Amazon’s Alexa recommends when you make your next purchase within that category. This repeat-buy mechanism poses a challenge to brands that want to become a habitual purchase.

In an effort to “hack” Prime Day, Cheerios offered a free family-size box of Honey Nut Cheerios to any Amazon shopper who spent more than $40 on Amazon Pantry. In addition to the box of cereal, customers received $10 off their order. The offer automatically created a Cheerios shopping history for millions of Amazon Prime shoppers, infiltrating Amazon’s recommendation algorithm and moving Cheerios to number one in the cereal category.

The Results? Cheerios became the #1 grocery item on Prime Day and Y-O-Y sales of Honey Nut Cheerios increased by 64%compared with the week before the activation. The campaign also claims that 80% of those who bought Cheerios in July were new to the brand.

Why it’s hot? More households are adopting voice assistants like Alexa and Google and with that, voice-assisted shopping is on the rise. Although voice-assisted shopping is still far from replacing online shopping,  brands need to get creative and test the limits of these platforms in order to capitalize on the voice-assisted retail boom.

Cheerio’s clever Alexa “hack” not only served as a mass-sampling campaign, but also ensured future purchases.
Source: Contagious

At Unilever, Resumes are Out – Algorithms are In

The traditional hiring process for companies, especially large organizations, can be exhaustive and often ineffective, with 83% of candidates rating their experience as “poor” and 30-50% of candidates chosen by the company end up failing.

Unilever recruits more than 30,000 people a year and processes around 1.8 million job applications. As you can imagine, this takes a tremendous amount of time and resources and too often talented candidates are overlooked just because they’re buried at the bottom of a pile of CVs. To tackle this problem, Unilever partnered with Pymetrics, an online platform on a mission to make the recruiting process more predictive and less biased than traditional methods.

Candidates start the interview process by accessing the platform at home from a computer or mobile-screen, and playing a selection of games that test their aptitude, logic and reasoning, and appetite for risk. Machine learning algorithms are then used to assess their suitability for whatever role they have applied for, by matching their profiles against those of previously successful employees.

The second stage of the process involves submitting a video interview that is reviewed not by a human, but a machine learning algorithm. The algorithm examines the videos of candidates who answer various questions, and through a mixture of natural language processing and body language analysis, determines who is likely to be a good fit.

One of the most nerve-wracking aspects of the job interview process can be anticipation of the feedback loop, or lack thereof – around 45% of job candidates claim they never hear back from a prospective employer. But with the AI-powered platform, all applicants get a couple of pages of feedback, including how they did in the game, how they did in the video interviews, what characteristics they have that fit, and if they don’t fit, the reason why they didn’t, and what they believe they should do to be successful in a future application.


Why it’s hot:
  Making experiences, even hiring experiences, feel more human with AI – The existing hiring process can leave candidates feeling confused, abandoned, and disadvantaged. Using AI and deep analysis helps hiring managers see candidates for who they are, outside of their age, gender, race, education, and socioeconomic status. Companies like Unilever aren’t just reducing their recruiting costs and time to hire- they’re setting an industry precedent that a candidate’s potential to succeed in the future doesn’t lie in who they know, where they came from or how they appear on paper.[Source: Pymetrics]

The list of Short-form video content producers just got longer

Last month, former Dreamworks film exec Jeff Katzenberg and former CEO of HP Meg Whitman finally announced a name for their stealthy new startup – Quibi (for “quick-bites”). Quibi is the latest addition to the increasingly crowded streaming video space, offering short-form video content designed specifically for mobile. While streaming services like Netflix, Hulu, and Cable Networks continue to battle-it-out for our 2 hours of attention on their long-form video content, Quibi believes they have identified a new niche of consumers in the rapidly emerging short-form video content space. Short term video content now consumes an average 70 minutes of our attention a day and growing, and now Quibi is betting that 20 minutes of that time will be spent on its platform.
While platforms like YouTube, Instagram, and Facebook were built with and continue to thrive on original user-generated content, Quibi is promising users the quality they’d get from a big Hollywood production in the form of the short, bite-sized content you’d find on Tik Tok, Snapchat, etc. Traditionally, short-term video content is less than 60 seconds, but Quibi wants to take what would be, for example, a 2-hour feature film and unfold it over several multi-minute chapters. Imagine sitting down to watch a movie and only being able to watch eight minutes of it at a time. The platform promises to publish more than 100 pieces of this type of content every week, including both scripted and unscripted original content, exclusives from Quibi’s partner, and other daily news and sports programming.
With existing platforms like Netflix and even Amazon are adapting to this desire for short-form video content, success may seem like a long shot, but Quibi has already managed to recruit a long list of high-profile partners, including filmmakers Sam Raimi, Guillermo del Toro, mega-pop musicians Justin Bieber and Justin Timberlake, and even basketball legend Kobe Bryant. The streaming service isn’t set to launch until April 2020, however, the platform is stirring up conversations of the future of TV and how we digest content.

Why it’s hot: As our daily lives become increasingly more busy, and with more platforms than ever competing for our attention, Quibi is taking a huge gamble on the future of TV and the future of wholly-owned short-form video content. With our limited attention spans and on-the-go lifestyles, there’s a growing need for platforms to adapt and change to how we digest content throughout the day.