Downside of AI, affecting society in all aspects

While we are certainly far from producing a Skynet that could wipe us all, there is no doubt that AI would be a double-edged sword. A latest 100 page report written by 26 AI experts have warned us of the potential threat AI imposes on the world.

The development of smarter AI might increase the creation of fake news or information. Currently, developers are creating AIs that could write natural articles and also AIs that could alter video content without viewers noticing the difference. If people with evil intentions were to misuse such technology, they would be able to produce deceiving and harmful content that might have detrimental impact on society.

AI might also change human warfare completely. With the combination of combat AI and advanced robotics, countries could develop killing drones or robots that have absolute accuracy, high mobility and infinite stamina.

The world is joining hands to limit the downside of AI

With such threats in mind, different parties of the world are already taking measures to reduce the downside of AI.

The United Nations is taking the initiative to develop protocols for the restriction of killer robots. Meanwhile, philanthropists like Elon Musk and Sam Altman have invested much to research how to ensure the safe use of AI. Despite of growing concern, there are only around 100 researchers in the West focusing on this topic.

While leaders are working together to formulate practical means to reduce the harmful effects of AI, there are a few issues that the world must discuss and decide on. For example, should tech companies be held responsible for the misuse of their AI products? Should there be guidelines on what is a qualified and safe AI product to be put into the market?

These questions and concerns will take much time to be answered. It is true that we can never uninvent a technology, so let’s hope that the world is working together to contain the downside of AI.


Upcoming AI Product: Microsoft + Xiaomi = ?

The world has witnessed the amazing potential of AI last year. From Alpha Go to autonomous vehicles development, 2018 will certainly be another great year for AI. Not only will AI technology be used by professionals from the business, medical, education and other sectors, it will also be available to the general market.

Cortana, an AI digital assistant developed by Microsoft, was released in 2014. As a virtual assistant, Cortana could comprehend natural voice demand and remind users of their schedule, weather forecast and news update. Available in all Windows 10 systems, Microsoft took Cortana to the next level by developing a smart speaker in late 2017. The smart speaker, Harman Kardon Invoke, however, struggled to compete with its competitors like Amazon, Apple and Google.

Collaboration of Microsoft and Xiaomi, budget AI speakers and Smartphones

Earlier this month, Microsoft and Xiaomi signed a Memorandum of Understanding which talks about their potential collaboration to develop Artificial Intelligence-powered smartphones and speakers. Microsoft would provide the Chinese IT giant with their cloud computing products such as Azure and Cortana. Meanwhile, Microsoft would gain more influence on the Chinese market from the partnership.

Xiaomi, the fifth leading company to produce smartphones and other gadgets has paved its way to success by producing affordable products. Potential products to expect include AI-driven smartphones, speakers, laptops and other gadgets. The two tech companies are also discussing other projects. Microsoft may offer Xiaomi other AI related technologies such as Skype, Edge and Bing. While the Memorandum of Understanding is not legally binding, Wang Xiang, Senior Vice President of Xiaomi, has tweeted about his confidence and enthusiasm about the cooperation “This will help accelerate our pace to bring more exciting products and services to our users, and also allow Microsoft to reach more users using Xiaomi products!”


AI: The tech double-edged sword

Technological singularity, is the hypothesis that once advanced artificial intelligence is developed, humanity would be revolutionized in all means imaginable. Some scientists predict that 2029 will be the earliest year for singularity to occur. While we can’t tell if AI would pass the Turing Test in 2029, we can be certain that AI will be a game changer in coming decades.

However, like any other revolutionary invention, AI is a tool that could be used for both good and malicious purposes. Artificial Intelligence Experts have written a 100-page report to warn the world about the harm AI could bring when fallen into the wrong hands.

AI : More fake news?

Currently, AI can already produce seemingly genuine articles and videos. Some developers have created AI that could produce fake news articles by studying different news websites. There are even AI programs that can change the faces of individuals into another or alter the weather in a video. Though they were for academic purposes initially, the report warns that AI will become a tool for generating propaganda. For example, fake images and videos can take advantage of publicly-available information to alter people’s opinions.

AI: Faster hacking, more information lost

The report warns that AI hacking systems could discover and attack loopholes in websites and softwares automatically. It can also target multiple systems simultaneously and efficiently. Since AI is self-learning and ever-evolving, it would be highly difficult for data security specialists to defend their system quicker than AI. We could expect that the future cyber warfares will be conducted by AIs instead of humans.

AI: Advanced Warfare/Terrorism

While it is unlikely for us to see a T-800 in the warzone in the future, AI would change the military arena greatly. AI will be able to pilot advanced drones equipped with lethal weapons or other military vehicles to perform strategic missions. The US has even developed MAARS, an autonomous tiny tank equipped with M240 machine guns and grenade launchers. The birth of real life killer machines will spark many ethical controversies. Meanwhile, if advanced technology falls into the wrong hands, the consequences are unimaginable.



Get ready for a great development of AI in 2018

Since the victory of Alpha Go, the world has been amazed by the tremendous potential of AI technology. Finance, business, aviation, healthcare and many more sectors have implemented AI, machine learning and deep learning into their daily operations.

Bill Gates recently said that AI will be beneficial to our society. AI is no longer merely a bot to mine data or predict trends. Instead, it has become involved in many different human activities. Here are a few AI trends to look out for in 2018.

2018: The year of autonomous (AI) robotic workforce

There will be more AI programmed robots working in the assembly lines of factories. As the technology of AI and robotics become more mature, autonomous robots will be able to perform more complex tasks efficiently. It is predicted that by 2025, more than eighty percent of elderly care service will be performed by robots.

2018: AI will be present everywhere

In 2018, it is expected that AI will become more versatile. In the past few years, we can see that finance and medical sectors have developed AI heavily. It helped with investment in businesses and also daily operations of medical professionals. In 2018, AI will be impacting many other fields. For example, AI with deep learning ability has been developed to identify fish for  eco research. AI will also soon be made to analyze behaviours of horses using machine learning and deep learning algorithms.

2018: AI talents are scarce

There are around 300,000 talents in the field of AI in the entire world. As more corporations are trying to develop AI programs for their businesses, they will have to compete against each other to recruit these AI talents. Currently, some positions in the field of AI have 6 digits salary. It is expected that companies will be willing to pay even more as AI becomes vital to business development.



Artificial Intelligence for Business Professionals: AI, Machine Learning and Deep Learning

Since the birth of Alpha Go and other successful AI projects, AI has become the hottest topic in both the tech and business arena. Suddenly, everyone is talking about AI. From AI marketing to AI financing, the discussion of AI has been so overwhelming that it is difficult to tell how a company can truly make use of AI for their business advantages.

AI, machine learning and deep learning are probably the most common words you hear from business professionals. The three are the fundamentals of the current AI trend. Without a complete understanding of the terms, implementation of AI will not be successful. This article will walk you through the meaning of these three terms.

Artificial Intelligence

AI is certainly not a brand new concept. It first appeared after WWII during which the academic community was pursuing a machine that could replicate the human mind. The famous Turing Test developed in the 1950 was an example of the world’s expectation of AI’s potential. As time goes by, scholars have found that it is highly difficult for humans to replicate the way our brains function.

Although scientists have yet to develop an AI that can think like humans, there have been other advancements in recent years. For example, AI is developed to read and understand research articles for a busy Professor, doctor or lawyer while providing them with a summary. Instead of copying the human mind, AI acts as a tool to enhance it.

Machine Learning

Similar to AI, machine learning is not a new concept. It refers to the computing skills that allow a program to predict and analyze certain trends. For instance, if a person has a series of data that they can correlate with significant events, they can form a pattern. By teaching machines about such patterns, they can then help predict and alert the user when similar trends appears. Machine learning is a common tool for businesses to analyze market movements and developments.

Deep Learning

Compare to the other two, deep learning is developed much more recently. Some scientists would refer to it as neural network because its concept originates from neurons and synapses within the human brain. A program/machine with deep learning ability no longer conducts linear calculation but multi-dimensional ones. Some examples of its application include programs that can automatically generate handwriting or one that can add sounds to silent movies on its own.

AI + Machine Learning + Deep Learning

In short, the current trend in the tech arena is all about the combination of the three. AI, ML and DL are interrelated. For businesses to truly excel in the AI era, they must develop a customized combination of the three technologies to fit their business needs.



With AI, medical professions are saving lives, more than ever

With the development of AI, machine learning and deep learning, scientists are applying computing technology to the medical field to save thousands of lives.

People can now use a smartphone to detect cancer, and a smartwatch can identify signs of stroke. This cutting-edge technology is becoming more versatile and can be easily applied with the flick of a finger.

AI to enhance medical professions

In Copenhagen, an AI software called Corti has been developed to provide real time information for emergency service dispatchers. It analyzes conversations and provides dispatchers with on-screen medical information. The information has proved to be effective for accurate diagnoses even in a critical and hectic situations.

Such collaboration is effective because AI is not used to replace humans but to enhance their work instead. Unlike an autonomous system, Corti and medical professionals work together. While the AI provide the dispatchers with relevant and vital information, it is always the human who makes the final call. As such, Corti has proven to increase citizes’ chance of survival in critical circumstances.

With open source, AI is going EVERYWHERE to save lives

An AI program called RadiO has been developed to aid cancer detection. With deep learning, it helps doctors look for lung cancer symptoms in radiology scans. The greatest perk of RadiO is that it is open source. It is available for anyone in the world to be used.

With the AI algorithms provided, doctors and medical professionals will be able to detect lung cancer quickly and accurately. The processed data will be used to further enhance the usefulness of the AI software.

AI in the medical field, an ideal development model?

While there have been predictions that AI will replace a number of jobs, the development of AI has shown a different possibility. Although powerful, AI will perhaps assume an assisting role in the future.



AI and Machine Learning in Education

The world has witnessed AI in the financial, aviation and medical field. Dubai is bringing AI into the education sector. In the UAE, education institutions like universities and colleges are applying AI and machine learning to predict students’ employability and their possibility of dropping out.

Nexsquare, a start-up founded in 2015 is a pioneer in the field. Cooperating with education authorities and Universities in Dubai, Nexsquare has developed programs to help improve education delivery and management.

AI is helping education institutes predict students’ performance

By analyzing the socio economic backgrounds, academic results and attendance of students, AI could predict the pattern of a student’s performance. It will then calculate the possibility for a student to underperform or to drop out. Teachers and the schools would then be able to take precautions before the situation deteriorates.

The same technique can be applied to recruiting academic staff. By analyzing similar datas, AI could let recruiters know how likely would a candidate perform well and how likely a current teaching staff would leave the institution.

The developer of the AI program stressed that the purpose of these predictions was never to decide a student’s fate. For instance, a school would not terminate a student’s study even if it foresees that he or she will underperform. Instead, it alerts the teaching staff and the school so that they can reach out to the student and work things out together.

Education with AI, not just a fancy gimmick but a practical development

The education authority of Dubai is pushing universities to adopt the tool. They believe that AI is not simply a fancy tool. Instead, educators will have to shoulder the responsibility and incorporate AI technology into their everyday operations. AI will enhance institutes’ evidence-based decision making and help them deliver better education.


AI: What’s Next? On Demand AI Means OnDemand Intuitive Marketing

aima manifest.jpeg

It’s not hyperbole to state that all of us — our behaviours, buying decisions and ultimately our thoughts — are constantly informed by a cascade of clever algorithms that have learned our patterns.

It’s an absolute certainty that those of us leading marketing will continue to work diligently to apply machine learning algorithms to the myriad of real world items listed in Federico’s Artificial Intelligence Marketing Manifestojust one year ago. These included Predictive Customer Service, Dynamic Product Pricing, Forecasting, plus sophisticated Image, Voice and Language recognition algorithms.

Federico, who founded the Artificial Intelligence Marketing Association(AIMA) just 1 year ago (Jan 2017) has seen the San Francisco chapter grow and expand, attracting thousands of local followers and magnetizing AI marketers (270+ in the bay area) across borders to the point that we’ve attracted 3 international conferences: The AI CongressBusiness of Botsand AI Expo as partners. Also, he will be speaking at the upcoming AI Congress in Las Vegas in early May presenting “The State of AI Marketing”. As well as at the Business of Bots conference in San Francisco.

In addition, AIMA SF will BE rolling out a series of Meetups in San Francisco on the intersection of advertising and artificial intelligence.

What’s Next for AI-empowered Marketers? Decentralization…And it’s Already Here

The 3 main obstacles companies have always encountered when implementing AI to uplevel marketing results are:

  • Data — which at best is costly — in fact so much that both Fortune and The Economist have declared “data is the new oil”. Beyond the ever-increasing high cost of data, it is also notoriously flawed.
  • Analysis — to ensure the data is properly interpreted.
  • Machine Learning — believed a solution but still an obstacle. It can create new, smarter algorithms based on existing datasets, in order to deliver actionable, revenue-producing insights — such as micro and macro consumer predictions to pinpoint how and what advertisers and marketers should message out next.

Taken 1 by one, first there’s Data. Every marketer knows the obstacle of Good Data.

If you have the massive reach of say, Coca-Cola, then you have the budget and data to deliver predictions and bulls-eye marketing messages to discrete target audiences.

But if you’re an advertising or marketing professional within a smaller agency, you and your teams are beset with this first massive obstacle; let’s face it, the financial distraction of data buying eats into team focus and profits.

Same with Analysis. Every advertiser knows that there’s a reason why Data Scientists and Data Analysts are the hottest jobs right now, and stumbling blocks this massive data talent shortage presents to advertisers and marketers dependent on data gathering and interpretation.

Finally there’s Machine Learning. It’s worth noting that machine learning can function as both a solution and an obstacle. The reason: Buying, testing and applying the right algorithms to accomplish advertisers’ ever-pressing need for insights is costly and can be time-consuming. But the machine is also the solution that the advertiser can rely on to ensure their messaging remains custom, compelling and intelligent when the technology is available and easy to use.

But Thanks to Blockchain Innovators, these 3 main Obstacles are Rapidly Disappearing. Here’s What’s Next…

Several AI companies are at the forefront of removing these obstacles — or to use a Silicon Valleyterm — “disrupt” these hurdles to the point that by end of this year we expect this Terrible Trio (Data, Analysis, and Machine Learning) to all but disappear.

How do they do that?

Blockchain Innovators Are Doing It Faster, Better and Cheaper — Right Now.

Companies like Synapse.AI and have been hard at work creating new platforms and new ways of making data cheaper, interpreting datasets in the right way and making machine learning algorithms accessible to everyone, e.g. to future AI-empowered marketers.

According to Zurich-based, a decentralized marketplace that allows AI algorithms to cooperate and coordinate, “when data is more accessible marketers can create, monetize, and use AI at scale,” said Dr. Ben Goertzel, founder.

Dan Gailey, CEO of San Francisco-based (currently headed into ~$7M ICO) is set to empower advertisers, marketers — anyone really — to do just that.

Gailey is creating an entire free, open AI ecosystem which essentially functions as the Wikipedia of AI algorithms. Gailey envisions a wide open repository containing fresh data, new algorithms based on innovative machine-learning tools through which marketers (and anyone, really) can readily gain the insights they need by simply creating a free account.

The future will be built on a fair and balanced intelligence available to all equally, by utilizing newly-created AI economies that any agent can participate in and receive some reward for doing so”, said Gailey in a February announcement about the public token sale.

Adding to the decentralized data disruption is, an ecosystem for small and medium-sized enterprises (SMEs) that will enable them to monetize on their data by selling it to 3rd party developers. Developers will then create new applications based on SME’s data and then offer their applications for sale into the platform.

The net-net is that the data cost and availability obstacles — which are currently the biggest obstacles — will diminish or evaporate. The net-net is empowering marketers to more readily and inexpensively deliver predictive analytics services to improve SMEs’ business processes.

“The monetization opportunities for SMEs and developers are huge; at this point in point in time, data services providers like AliBaba, Salesforce, Google represent $4T worth of services”, according to David Siu, spokesperson.

Think About It:

  • What will the elimination of current the 3 major AI barriers mean for marketing professionals?
  • What will it mean when data and machine learning tools are free and ready to access as easily as logging into Gmail each day?
  • What will it mean for marketers on the front lines to have readily-consumable answers at their fingertips? To be able to log into a free repository where a simple search turns up fresh data and applicable answers?

Put yourself in the shoes of a frontline marketing professional at a small or medium-sized business. Your CEO tasks you with delivering predictions for Q3 advertising messages based on data gathered in Q1 & Q2. Prior to the intersection of data sharing and blockchain, this task would typically have represented a cost — cost of gathering the data, cost of purchasing 3rd party software to interpret the data and others, in order to arrive at the critical insights required.

Now with the advent of new AI innovations like Synapse, RepuX and others, the cycles of accurate predictive planning will be unfettered.

How this will impact the future of big data providers who currently make millions in revenue by selling customer information?

Well…we at AIMA predict this will do to the big data sellers — — and perhaps to data analysts — what Google did to Encyclopedia Brittanica salesmen. And map sellers.

Concluding, soon marketing departments — and not just them — of SMEs will be able to implement big data providers information shared on the new blockchain infrastructures, as well as access AI-drag-and-drop algorithms (AI for Dummies) on top of their current marketing stack and gain fully AI-empowered marketing heroes — — and perhaps without the need of data analysts, we’ll see…. !

AIMA will be presenting at the upcoming AI Congress in Las Vegas in early May — we hope to see you there!

aima twitter.PNG

If you can’t make to Vegas, we’re rolling out a series of Meetups here in San Francisco — join us to hear top insights from advertising campaign leaders and marketing pros about how to stay ahead of the increasing intertwining between marketing and artificial intelligence.

How AI will help with the HR process and what it means for your business

Ever since the success of Alpha Go and various other AI projects, the world is certain that Artificial Intelligence is the next big thing. The tech, finance and government sectors all know that 2018 will be the a huge year for AI. A report by Tractic even predicts that the revenues from AI will be $59.8 billion by 2025.

AI: Minimizing the toil of HR process

One major reason AI is generating huge revenue is its potential to reduce the workload for employees, making their duties easier. AI is fundamentally faster and more efficient than humans as it is a self-learning and automated process.

With the development of language processing AI, AI will help scan and identify the top resumes of applicants. In the past, HR managers often had to go through all the CVs manually. They either spent lengthy hours studying all the details or they screened out certain candidates based on some benchmarks. Both of them are not ideal for a holistic recruitment process. AI, however, will be able to analyze both quantitative and qualitative data and pick up the best match for the hiring manager. It can even adapt to the preference of the HR department and become more accurate as it accumulates more experience. Screening and identifying resumes will no longer be a tedious job for HR managers.

AI: Handling the screening interviews, on its own

Most corporations have a few rounds of interviews. Let’s be honest, it is always the final interview that seals the deal, the preliminary ones are for screening out the odd ones only. Still, companies often spend quite a lot of time on the initial screening process and it could be inefficient and time consuming. AI will be able to handle these screening procedures on its own. With the help of facial recognition programs and AI, machines will be analyze the facial expressions, attitudes and competency of candidates via videos.

AI: Improve Efficiency of Recruitment, doing what we couldn’t

Not only will AI handle tedious tasks in the future, it will also perform complex tasks beyond human capacity. AI will be assessing candidates by understanding their activity on social media and performing personality analysis. By looking through candidates’ social media profile, AI will highlight the candidates’ outlook and even missteps in their lives. It will alert the HR Manager for inconsistency between candidates’ words and personal life. Candidates will have to make sure they truly are the image they are presenting to the employers.

AI in HR, any catch?

While it sounds perfect to have an AI to aid your recruitment process, companies must develop their own algorithms and constantly monitor the performance of the machine. Afterall, HR is a human process.


Facebook Appoints Former IBM Watson Executive as Head of AI

Facebook hired a former IBM executive to spearhead the company's artificial intelligence ventures.

The shift in focus is also evident in Amazon, Google and Microsoft, who are likewise bulking up their AI departments.

Jérôme Pesenti, who headed development of IBM's Watson platform, has now been appointed as vice president of AI, a Facebook spokesperson told CNBC. More recently, Pesenti worked at the BenevolentAI, according to Quartz.

At the samae time, Yann LeCun, AI expert and an alumnus of New York University will be promoted as Facebook's chief AI scientist following his previous position as the director of the Facebook Artificial Intelligence Research unit.

A few leaders of FAIR group, as well as the head of Facebook's Applied Machine Learning Group, will work under Pesenti, who in then report to Mike Schroepfer, Facebook's CTO.

Facebook will grow its AI capacities under Pesenti, explained the spokesman. The FAIR group comprises of over130 people globally. Besides equipping the team for growth, the shift will also assist the technology transfer from the FAIR to AML group, said the spokesman.



Baidu Jumps on the AI Bandwagon

“Now that AI has become mature enough to have some use in the market, we feel relatively more relaxed,” said Robin Li. 

Baidu Chief Robin Li, said that they won't pass on artificial intelligence after their search engine operator made the mistake of not understanding the importance of mobile internet back in the early 2000s. 

The share price of Baidu, also known as “China Google”, has been rising yet again as investors are noticing that the Beijing based company is investing heavily in artificial intelligence. 

Back in 2002, Li was unprepared for the coming of web access centred on the mobile phone and he doubted the potential and possibility of the success of mobile phones due to small screens, slow speed and the leading edge payment method. In 2011, Li started to panic as main rivals Alibaba Group and Tencent Holdings had already positioned themselves in the mobile internet era, while they were not. 

Nonetheless, Baidu sets its eyes on being a leader in artificial intelligence by 2030. The goal for Baidu is to build an open-source platform for autonomous driving to compete with those being developed in traditional monitoring powerhouse like the US, Japan and German. 

Baidu's Vice Chairman and Chief Operating officer, Lu Qi, claimed that with government backing and China's giant population, they are well positioned to catch up with the US in the field of artificial intelligence. Baidu made several major changes, such as increasing its efforts in autonomous driving and its conversational computer platform DuerOS.



Microsoft Introduces New AI Drawing Bot

Microsoft recently introduced a new artificial intelligence (AI) software – “Drawing Bot”. Drawing Bot is an artist, able to draw variety of things and turn something from text description to vivid images. It is able to convert text details into lifelike images, showing that AI itself has some imaginative power, says Microsoft. 

“If you go to Bing and you search for a bird, you get a bird picture. But here, the pictures are created by the computer, pixel by pixel, from scratch,” explained Xiaodong He, principal researcher and research manager in the Deep Learning Technology Center at Microsoft’s research lab in Redmond, Washington. “These birds may not exist in the real world — they are just an aspect of our computer’s imagination of birds.” 

The software has a well developed dataset which enables it to pair up descriptions and images, Microsoft explained. For example, when the drawing bot is told to draw a bird in text, it pictures how a bird should look like.

Two machine learning models are incorporated into the drawing bot. One is for translating text descriptions into images, and one for determining the reliability of the produced images. The producer tries to get rid of fake images and put two compatible models together in order to deliver better and higher quality images.

According to recent research, this new bot is almost three times better in generating quality images than past past technology.

Microsoft’s AI drawing bot is not the first software to integrate with art. Google, Facebook and some leading big brands are also adopting AI to help with drawing and producing images.

The AI bot could be a useful tool in different job areas, such as animation. It can down on manual labor in the production of animated movies. However, this technology still requires time to continue developing. 


Microsoft Drawing bot.jpg

Wal-Mart Trials AI Robots

It is never a fun job for employees to spend hours and hours keeping track of inventory, especially in a huge company like Wal-Mart Store, which has more than 20 million items sitting on the shelves. 

However, this job must exist. Without it, the retailer will suffer from the loss of profit from every out-of-stock item. Hence, the cost incurred will be even higher. This is especially important to online stores, considering the increase of online shopping on platforms such as

Wal-Mart came out a clever idea to solve this issue by adopting an autonomous robot started up by Bossa Nova, which is combined with artificial intelligence (AI). 

The height of the robot is just two feet tall, and it comes with an extendable “foot” that allows it to reach above six feet. It drives itself around the store and captures the shelves in photos which are then stored in its memory. In only two minutes, the robot can inspect up to 80 feet worth of units in the store. It can then pick out the out-of-stock products, spot wrongly priced items and identify any labels that are missing.

The robot can store two terabytes of data. It uses AI to analyze raw data and calculate the status of each item on the shelves in terms of its price, location and availability. The processed information will then be passed to Wal-Mart employees to take further actions if required.

"Access to data is the foundation of a truly seamless omni-channel retail business," said Martin Hitch, chief business officer at Bossa Nova. "Our solution provides visibility into on-shelf inventory and helps the retailers improve store operations to better serve their customers, whether in-store or for online pickup. Wal-Mart is retail's innovation leader and we're excited to be a part of their advanced technology initiative."

Humans are not as efficient than AI robots when doing this kind of job. Robots are 50 per cent more productive when it comers to searching for items for customers. They can complete the task three times faster than humans, and they are even more accurate when it comes to this, Wal-Mart's Chief Technology Officer Jeremy King stated.

Wal-Mart is expanding the assessment to 50 other locations after limiting trials to a few stores in Arkansas, California and Pennsylvania. The location and time allocated to robotic assistance will be decided based on customer feedback.

Nowadays, many other retailers are also using robotics in store to enhance their working operations. Leader of E-commerce Amazon spent $775 million on Kiva Systems. The orange robot keeps track of the shipping centers and completes all the shipping processes efficiently, from selecting the correct stocks to package to getting them ready for shipment.



LG Infuses AI Into ThinQs Gadgets

At this year’s Consumer Electronics Show in Las Vegas, LG focused solely on artificial intelligence for the majority of its presentation. 

During a major press conference at the CES, LG demonstrated how its washing machines can adjust cycles depending on the weather, air quality and the exercise schedule of their owners through AI. It also explained how their air conditioners would know when to turn itself off and on according to their owners’ habits. In addition, its robot vacuums would be able to tell apart householders and furniture.

Most importantly, LG announced a highly anticipated deal with Google. Google Assistant would be included in LG TVs at the end of 2018 and allow users to command household appliances remotely through voice control on the TV.

Whilst the details of LG and Google’s collaboration have yet to be revealed, it is very likely that Google’s Assistant will work with LG's own AI platform ThinQ instead of replacing it. From the new smart TVs’ demo video, users communicated with ThinQ, which then bounced content-related requests to Google Assistant.

The integration of multiple AI platforms represents one of the main challenges within the consumer electronics industry. As such, what LG and Google are doing could potentially be groundbreaking, as they are working towards to streamlining separate, and even competing systems.

Another major obstacle relating to AI is the difficulty of making it seem more authentic and smart. This was evident in LG’s setback during the press conference.

CLOi, a robot that is designed to provide domestic help, did not respond to David VanderWaal - LG vice president of marketing - when asked to set the LG washing machine to a personalized setting suitable for a gym fanatic. 

CLOi’s failure to respond to Mr VanderWaal’s command raises concern on technology’s reliability, not only at public demonstrations such as CES, but also more widely in everyday life. 

Many who work in the AI industry understand that the incorporation of AI into household items is still in its formative stages. We should be prepared to face failure, but remain hopeful for what the future holds for AI technology. 

Source: Financial Review



Facebook Pushes A.I. Beyond M Assistant

With stiff competition from Alexa, Cortana, the Google Assistant and Siri, Facebook needed to create an assistant who can stand out from the rest. Facebook introduced M in its Messenger chat app in 2015.

Different from generic chatbots that can automatically generate text responses from message histories, M Assistant is capable of handling most tasks, as long as they do not breach the law. However, its use was limited to only 2,000 people out of 1.3 Messenger users.

Last year Facebook added automatic actions and responses into Messenger conversations for millions of users. Shortly after, they got rid of the human-powered version of M once and for all.
Considering Facebook's broad deployment AI, M assistant pales in comparison to other features. Facebook counts on AI to interpret text, photos and videos for billions of users daily.

Below is a list of ways in which Facebook uses AI beyond M assistant:

  • Predicting response to messages on Facebook Messenger
  • Blocking offensive and inappropriate comments on posts
  • Stabilizing photos and videos
  • Location tracking for friend and event recommendations
  • Facebook face recognition and notification in photos
  • Automatic translation of texts and captions
  • Adding interesting filters and effects to photos and videos
  • Compiling relevant content on ‘Anniversary’ and ‘Friendship’ videos
  • Detecting and purging content distributed by terrorists
  • Identifying posts and Live videos that hint at dangerous behaviour
  • Determine which posts to show on a news feed and in what order
  • Showcasing the best ads to Facebook users
  • Generating content under the search function

Having established its AI research lab back in 2013, Facebook has not yet ceased its search for new talent. As such, it is apparent that the company is determined to continue growing its AI capabilities.

Source: CNBC

FB AI.jpg