Xinhua, the Chinese state news agency, has released its latest artificial intelligence (AI) 3D news anchor. The AI anchor joins a list of growing virtual presenters that are being developed by the agency.

The AI news anchor is named Xin Xiaowei, and it is modeled after Zhao Wanwei, who is one of the agency’s human news presenters. 

According to the search engine Sogou, who co-developed the technology, the AI anchor utilizes “multi-modal recognition and synthesis, facial recognition and animation and transfer learning.”


The video released by Sogou shows Xin Xiaowei speaking on set about how the anchor can “intelligently imitate human voices, facial expressions, lip movements and mannerisms.” 

Previous Virtual Presenters

Xin Xiaowei is not the only virtual presenter that has been developed by Xinhua and the Beijing-based Sogou. It joins a growing list that includes their 2018 digital anchor Qiu Hao and 2019 Russian-speaking version

In 2018, the pair debuted two different AI news anchors, identical to each other in appearance, at the World Internet Conference. The two versions’ biggest difference was language, with one speaking English and the other Mandarin. 

Both of the 2018 models were based on Zhang Zhao, who was another human-anchor like Zhao Wanwei. 

In order to develop these first models, hours of video footage was used to replicate the movements, expressions, and other features of real-life anchors. 

According to a report released by Xinhua in 2018, “AI anchors have officially become members of Xinhua‘s reporting team. Together with other anchors, they will bring you authoritative, timely and accurate news information in Chinese and English.”

The 2018 AI news anchors were used on various distribution channels including WeChat, the TV webpage, Weibo, and Xinhua’s English and Chinese Apps. 

The Russian-speaking anchor was released at the St. Petersburg International Economic Forum 2019. It was developed through a different partnership than the other two versions, with Xinhua working with Russia’s leading news agency, ITAR-TASS. 

The announcement came as the two nations celebrated their 70th year of diplomatic relations. 

China's State News Agency Introduces New Artificial Intelligence Anchor

First Russian-Speaking AI News Anchor. Image: Sogou Inc. 

ITAR-TASS Partnership

ITAT-TASS is one of the largest news organizations in the world, consisting of a network of businesses, media organizations, diplomatic missions, and financial and research institutions. They have over 1,500 reporters present in more than 63 countries. 

“We are very excited to launch the world’s first Russian-speaking AI News Anchor,” said Xiaochuan Wang, CEO of Sogou back in 2018. “The development of the Russian-speaking AI News Anchor allows us to share the benefits of Sogou’s leading AI technologies with more diverse audiences around the world. As one of the world’s largest news organizations, ITAR-TASS is an ideal partner for Sogou, and we look forward to introducing this new AI News Anchor to Russian-speaking audiences.”

The Spread of AI Personalities

The newest AI anchor coming from Xinhua and Soguo highlights the increasing presence of AI-personalities, especially in the realm of media. The technology is improving at such a rapid rate that it will soon be undetectable when put next to a real-life human presenter. 

The use of these AI-anchors could dramatically alter the media landscape, but it is really just a part of the larger takeover of AI in the industry. Whether it is AI writers, news anchors, or some other use for the technology, it is going to become increasingly difficult to differentiate between what is human-based and what is AI-based. 


  • 85% of enterprises are evaluating or using artificial intelligence in production today.
  • 55% of all enterprises adopting AI today are using TensorFlow as their primary development tool.
  • 73% of enterprises with the most advanced AI adoption levels say supervised learning is the most popular machine learning technique (73%).
  • Human-in-the-loop AI models are considerably more popular among enterprises with advanced AI expertise compared to their peers.
  • Enterprise's enthusiasm for AI is growing, with 62% increasing their spending last year, according to a recent MIT Sloan Management Review study.


These and many other insights are from O'Reilly's recently published research, AI Adoption in the Enterprise 2020 available for download here (20 pp., PDF, free, opt-in). The survey is based on interviews with 1,388 respondents from 25 industries, with 17% of total respondents from the software industry. 30% of respondents are data scientists, data engineers, AIOps engineers, or their managers. 70% of all respondents are in technology roles. For additional information on the methodology, please see pages 2 and 3 of the study, downloadable here.

Additional insights from the study showing enterprises' growing adoption of AI include the following:


  • In 2020, AI is being adopted evenly across enterprises, with R&D leading all departments by a wide margin. O'Reilly's survey finds that enterprises are stabilizing their adoption patterns for AI across a wide variety of functional areas. Nine to twelve functional areas included in the survey have over 10% adoption. It's fascinating to watch how IT is adopting AI to improve ITSM performance for example. AI has the potential for redefining enterprises, making them more customer-driven, adaptive, and capable of generating and sharing intelligence faster than ever before. Guiding the transformation of an enterprise takes a framework that both can create knowledge while staying focused on the customer and provides an understanding of the entire customer journey. BMC's Autonomous Digital Enterprise (ADE) shows potential in this area, as its structure enables all departments of an enterprise to contribute and share AI-driven insights about customers and provide a transcendent customer experience. The following is a ranking of the functional areas of enterprises where AI is used today:


AI Is The Uncertainty Cure Enterprises Want In 2020


Supervised machine learning algorithms is the most popular machine learning technique in enterprises today. 73% of enterprises with advanced expertise in AI are making extensive use of supervised machine learning techniques to interpret, classify, and analyze the large data sets they've accumulated over years of operations. Enterprises who are evaluating AI are using deep learning techniques in their pilots, making this area of machine learning most popular with enterprises running AI pilots today.

AI Is The Uncertainty Cure Enterprises Want In 2020



  • TensorFlow continues to be the most popular development tool across all enterprises evaluating and using AI in production today. The O'Reilly research team found that 55% of all enterprises in 2019 and 2020 place a high priority on TensorFlow expertise, making it the most popular tool two years in a row. TensorFlow is integral to completing deep learning and neural network projects, further evidence of how enterprises are adopting AI to solve increasingly complex problems. The following is an analysis of the AI tools enterprises are using today: 


AI Is The Uncertainty Cure Enterprises Want In 2020



  • 53% of advanced enterprises using AI today say the greatest risk when building and deploying Machine Learning models is unexpected outcomes and predictions. The more experienced an enterprise is using AI, the more they're likely to anticipate unexpected outcomes and work on making models more transparent. The most advanced enterprises adopting AI today are also far more likely to include steps during model building to improve fairness, ethics, and limit or control biases. 


AI Is The Uncertainty Cure Enterprises Want In 2020



  • Of the 15% of enterprises who are considering AI, one in five or 22% says that a lack of institutional support is slowing down adoption efforts. The most significant barrier to overcome is changing a company culture that doesn't recognize the value of AI. Conversely, the most successful AI implementations are known for the strong support for senior management they receive. Closing the skills gap is the third greatest impediment to making progress with AI. Comparing bottlenecks holding back AI adoption across the entire survey sample versus enterprises who have reached a level of AI maturity shows how significant the skills gap continues to be. The O'Reilly research team found that selecting the right machine learning technique for the job has more than three-quarters (78%) of respondents selecting at least two of ML techniques, 59%, using at least three, and 39% choosing at least four.


AI Is The Uncertainty Cure Enterprises Want In 2020


Source: Forbes

Russian researchers have revealed that artificial intelligence (AI) is able to infer people’s personality from ‘selfie’ photographs better than human raters do. Also Read - Early Signs of Glaucoma Progression to Blindness Can be Spotted by AI

The study, published in the journal Scientific Reports, revealed that personality predictions based on female faces appeared to be more reliable than those for male faces. Also Read - Kazakhstan Woman Goes to Celebrate End of Lockdown, Falls To Death While Posing For Photo On A Cliff

The technology can be used to find the ‘best matches’ in customer service, dating or online tutoring, the researchers from HSE University and Open University in Russia, said. Also Read - Mumbai Students Develop AI Tool Which Could Help Diagnose Covid-19 Through Voice


Studies asking human raters to make personality judgments based on photographs have produced inconsistent results, suggesting that our judgments are too unreliable to be of any practical importance.

According to the study, there are strong theoretical and evolutionary arguments to suggest that some information about personality characteristics, particularly, those essential for social communication, might be conveyed by the human face.

After all, face and behaviour are both shaped by genes and hormones, and social experiences resulting from one’s appearance may affect one’s personality development.

Read Stephen Diorio’s explaining how we can realize the growth potential of artificial intelligence on Forbes :

Business leaders and investors universally agree that Artificial Intelligence (AI) and Machine Learning (ML) will transform their businesses by reducing costs, managing risks, streamlining operations, accelerating growth, and fueling innovation. The potential for AI to drive revenue and profit growth is enormous. Marketing, customer service, and sales were identified as the top three functions where AI can realize its full potential according to a survey of 1,093 executives by Forbes.


Sales organizations are dramatically improving sales performance by using algorithms to help with the basics of account and lead prioritization and qualification, recommending the content or sales action that will lead to success, and reallocating sales resources to the places they can have the most impact.

·        Marketers are looking for AI to fuel enormous efficiencies by targeting and optimizing the impact of huge investments in media, content, products, and digital channels.

·        And in customer service, AI is opening entire new frontiers in customer experience and success by applying NLP, sentiment analysis, automation, and personalization to customer relationship management. 90% of organizations are using AI to improve their customer journeys, revolutionize how they interact with customers and deliver them more compelling experiences.

To realize this potential to grow revenues, profits and firm value, businesses in every industry have announced AI focused initiatives. On average, investment in advanced analytics will exceed 11% of overall marketing budgets by 2022. Spending on AI software will top $125B by 2025 as organizations weave AI and Machine Learning tools into their business processes. In parallel, investors have poured more than $5 Billion into over 1,400 AI fueled sales and technology companies to meet this demand.

So far, the impact of these investments on growth and profits has not yet been transformational. Right now 70 % of AI initiatives are showing little or no return. And more businesses will struggle to realize the full potential of AI to grow firm value if their leaders don’t learn lessons from past transformations like the internet in the 1990s and cloud computing in the mid-2000s, according to Kartik Hosanagar, Professor of Technology, Digital Business and Marketing at the Wharton School and author of the influential book A Humans Guide to Machine Intelligence.  

What separates the AI projects that succeed from the ones that don’t often has more to do with the business strategies organizations follow when applying technologies than the ability of the technology itself to transform the business,” according to Professor Hosanagar. “Many of the problems are less about the tools and more about leadership. Most of the failures to harness the power of AI lies in human behavior, management understanding, and the failure to mesh algorithmic capabilities into organizations, business models and the culture of the business.”

Today most executives feel like the pace at which AI can be made successful has been overstated, and the challenges have been understated according to the Forbes survey. That is totally understandable based on the current level of acumen in the business community about AI and advanced analytics. But the perception of hype and speed is an education and skill problem. AI works today in many business applications. It’s more a matter of the managers tasked with harnessing the power of AI don’t have the experience and framework to understand it.  Just as a calculus class will move far too fast for a sixth grader to grasp, growth programs based on AI and ML will be far too advanced for the executives who define, direct, and fund their development and are ultimately accountable for the results they deliver.

“Algorithms are opaque to the average business executive and can often behave in ways that are (or appear to be) irrational, unpredictable, biased, or even potentially harmful,” continues Kartik. “It’s up to business leaders to shape the narrative, direction, and ways algorithms can -and cannot - impact work, customer relationships, and the way business creates value.”

Executives who allocate capital and the managers who will lead the AI transformation cannot afford to have a poor understanding of something so fundamental to business and the creation of value today. “Ignoring the problem because it’s complex is not really an option. AI-based algorithms are here to stay,” continues Professor Hosanagar. “To discard them now would be like Stone Age humans deciding to reject the use of fire because it can be tricky to understand and control”

To help bridge this knowledge gap, The Wharton School of the University of Pennsylvania announced yesterday the establishment of Wharton AI for Business (Artificial Intelligence for Business), which will inspire cutting-edge teaching and research in artificial intelligence, while joining with global business leaders to set a course for better understanding of this nascent discipline. The goal of AI for Business is to educate a new generation of business leaders with a deeper understanding of AI – its fundamentals, capabilities, use cases, risks and limitations – so they can align AI with their business strategies and effectively direct, prioritize and invest in applying AI in their unique business models.

A cornerstone of the launch is a 4 week Artificial Intelligence for Business online certification program for business leaders and professionals. The program is aimed at providing executives, managers, and business professionals in the fields of marketing, operations, automation, and analytics a competitive edge in the emerging field of AI analytics.

According to Hosanagar, one of the primary reasons Wharton launched the AI for Business initiative is because it can help managers avoid very common mistakes their peers make when they define, invest in, and deploy AI-led transformational initiatives. Specifically, managers leading AI transformation typically make the same set of mistakes:

·        They execute AI development in siloes isolated from the business, or outsource it entirely, instead of making it a core part of the business;

·        They treat AI led transformation as a separate strategy instead of using it to support their core business objectives and growth agenda;

·        They fall into a trust and transparency vortex in which they either trust AI tools blindly without truly understanding them, or not at all, because they don’t understand what is inside their “black box” algorithms.

Kartik is emphatic that today’s managers must learn from the mistakes of past transformations. “Today nobody denies the internet was transformational to businesses and created billions of dollars of shareholder value,” reminds Hosanagar. “But despite the huge hype and promise, it certainly did not start that way. If you look back at the dawn of the internet 20 years ago, almost every organization quickly set up an independent division to lead the transformation to digital. Most of these failed.”  Hosanagar cites the example of Kmart who in 1999 aggressively invested in - a separate division - ahead of most of their competitors, but failed because they did not stick with it long enough and did not integrate the digital division with the rest of their business. The company soon went bankrupt in 2002. “A siloed approach to transformation is a flawed strategy. Ask yourself how many businesses have independent divisions anymoreWhat eventually did succeed was to find ways to use the internet to augment and accelerate their core business strategy – simplifying ordering, improving customer services, and supporting omnichannel sales models.”

“In my 10 years of working with data science and AI strategies in business, I see executives tend to fall into two camps when it comes to applying AI to their business,” shares Professor Hosanagar. “They either don’t understand it but trust it. Or don’t understand it and do not trust it. Both are failed strategies. The key message here is leaders need to understand enough about how AI works to strategically align AI with value creation and make smart investment decisions.” Specifically, Professor Hosanagar advises managers leading AI transformation initiatives to:

·        View AI as a tool, not a strategic goal;

·        Take a portfolio approach to AI project that balances quick wins with fundamental process redesign;

·        Grow your talent base by both re-skilling existing employees and hiring new talent;

·        Focus on the long term by sticking with AI through inevitable early failures;

·        Be aware of new risks AI can pose and manage them proactively.

Every executive must have a fundamental understanding of AI as they look to advanced analytics as a way to scale their businesses,” according to Sajjad Jaffer, founder of Two Six Capital, a firm that pioneered data science for private equity. “Wharton is unmatched in its depth and breadth of research in the fields of Statistics and Analytics. Programs like Wharton's new AI offering are table stakes for next generation leaders as companies increasingly rely on large data sets, cloud computing infrastructure, and open source software to scale their businesses," continues Jaffer, who serves on the board of Wharton Customer Analytics and is a Wharton Senior Fellow. "Investment committees and company boards need to bridge the widening chasm that exists between sound business judgement and AI skills across industries and asset classes."

Christine Cox, the VP of Marketing Operations and Demand Generation at Ricoh USA echoes this sentiment. “Based on my 20+ years leading marketing and sales teams across financial services, telecom and technology, AI is only just beginning to break into the Martech stack of traditional brands, enabling hyper-personalization of the Customer Experience,” reports Cox. “As large organizations develop greater AI capabilities for driving customer acquisition and retention, we will see these organizations innovate faster, engage with customers in new ways and start to compete with the digital-native companies. Holistically, AI has catapulted digital marketing and digital sales in the last five years, and I expect AI will exponentially accelerate the research and response process for marketing and sales teams to address evolving buyer needs in the future. However, this won’t happen with technology and data alone. In my experience, the business leaders who work to truly understand the nature and capabilities of AI and advanced analytics will be the ones who will realize the greatest impact and value from this transformation for their respective audiences.”

Saurabh Goorha, a Senior Fellow at The Wharton School, reinforces Kartik’s advice that managers gain an fundamental understanding of AI and ML to make them aware of new risks AI can pose and manage them proactively. “Executives make significant decisions about how they should invest capital, resources and talent to realize the full potential of AI and ML technologies to transform their businesses,” relays Goorha, who has decades of experience in product managment in EdTech and MarTech. “These decisions should be an outcome of a grounded understanding of AI and ML starting with first principles: what are the business and functional problems that can be solved and measured with comprehensive data strategyAt the next level they must ensure their AI strategies are informed by a solid understanding of both the potential and risks of AI as well as the strengths and limitations of the underlying data fueling these programs.”

Source: CTO Vision

Special Operations Command is gung-ho about leveraging artificial intelligence and machine learning capabilities across its portfolio as it faces off against peer competitors and violent extremist organizations, according to top officials.

SOCOM Commander Army Gen. Richard Clarke noted that a number of his program executive offices are keen on the technology.

“Artificial intelligence and machine learning efforts are integrated into most of our major PEO programs, and we’ll continue this,” he said May 12 during a keynote address at the virtual Special Operations Forces Industry Conference, or vSOFIC, which is managed by the National Defense Industrial Association.

SOCOM is in a “war for influence” against its adversaries, making military information support operations, or MISO, ever more critical, he said. MISO was once known as "psychological operations."

“As we look at the ability to influence and shape in this [information] environment, we’re going to have to have artificial intelligence and machine learning tools, specifically for information ops that hit a very broad portfolio, because we’re going to have to understand how the adversary is thinking, how the population is thinking, and work in these spaces” to conduct information operations at a fast pace, he said.

“If you’re not at speed, you won’t be relevant,” Clarke added. “What we need is adapting data tech that will actually work in this space and we can use it for our organization.”

SOCOM has now stood up a joint MISO “web ops” center, he noted.

The push comes as extremist organizations such as ISIS may try to take advantage of the COVID-19 pandemic to boost recruitment using social media.

“Extremists will capitalize on the economic situation as more people become prone to extremism,” said Jordanian King Abdullah Il bin Al-Hussein during a guest appearance. Countering those efforts through international partnerships and enhanced cyber capabilities will be critical, he noted.

SOCOM’s interest in AI and machine learning isn’t confined to information operations. Acquisition Executive Jim Smith said the command is pursuing a wide range of applications for the command’s top priorities, to include: next-generation intelligence, surveillance and reconnaissance; next-gen mobility; precision fires and effects; biotechnology; hyper-enabled operator; and data and networks.

For next-gen ISR, sensor fusion will be critical as the command seeks to tie in information provided by unmanned aerial vehicles, cyber- and space-based capabilities, and other sensors, he noted.

“The problem is each one of those sensors takes an operator offline, so how do we use artificial intelligence and machine learning to get those sensors to interoperate autonomously and provide feedback to a single operator to enable that force to maneuver on the objective?” he asked.

For next-gen mobility, AI and machine learning could help drones and other robots navigate and perform tasks autonomously, Smith noted.

“Think of those small UAVs or your small ground vehicles and giving them enough artificial intelligence and machine learning to be able to be autonomous, so that they can clear a building or they can clear a tunnel and … freeing up your maneuver force to be much more effective and efficient,” he said.

The technology could also help special operations forces employ radio frequency countermeasures.

“Today the way we do that [is] we have a library of threat radar signatures … on board, and if you see one of those threat radars in our library we counter it,” Smith explained. “Wouldn’t it be great if we had some type of machine learning that identified anomalies in this space so it wasn’t just the threat radars we had loaded into the library that we thought we might see in that theater, but maybe it’s a new radar that we haven’t seen before or a radar that we didn’t realize was operating in that theater that we could identify?”

For its precision strike portfolio, SOCOM wants loitering munitions that can fly around until they need to be called upon to strike a target at the right time and place. It also wants counter-drone technology. AI and machine learning could be used to identify enemy drones and tip off defensive systems, Smith noted.

In the biotechnology realm, the command is working with an industry partner using AI and machine learning to study the long-term health effects on the brain of low-level blast exposure. Smith did not identify the company or provide more details about the initiative.

AI and ML also feeds into SOCOM’s “hyper-enabled operator” concept.

“What we’re talking about in this case is improving their cognitive overmatch at the edge,” Smith said. That includes the ability to analyze, synthesize and communicate information to enable warfighters to make better decisions.

Analysts at tactical operations centers are fed a lot of information, he noted. “It’s great information, but what’s happening right now is the analyst has to take that information, ‘productize’ it, gather it, think about it, put it into a format and then disseminate it,” he said. “It may get to our individual operator at the edge in a timely fashion and may be tailored to their situation, but probably not,” he added.

To mitigate the problem, SOCOM will be kicking off a new “automate the analyst” effort at the SOFWERX office in Tampa, Florida, in June, he said.

The potential for AI and ML to bolster data management and networks is obvious, he said.

“We’re committed to rolling out specific things in the near term,” Clarke said.

Clarke noted that the command is already using AI for tasks such as predictive maintenance, as well as enabling mission command on the tactical edge.

“Our forces are conducting battlefield operations every single night,” he said. “We have AI teams with them looking at those processes, recording them, pulling in the data and then fusing this physical and information environment so we can make faster decision-making going forward.”

Clarke said having personnel that understand and can use AI effectively will grow in importance.

“We still need guys that can kick down the door, that can shoot well, can jump out of airplanes, can fly our special operators,” he said. “But we also need coders. We also need leaders who can apply AI.

“We’ve been having discussions internally that it may no longer be that the most important person on the mission is actually the special forces operator who’s kicking down the door, but it could be the cyber operator that the special operations team actually has to get to the environment and make sure that he or she can work his or her cyber tools into the fight,” he added.

Smith noted that the command has a new program executive office, PEO SOF Digital Applications, that will pursue AI software capabilities.

Across the acquisition portfolio, SOCOM wants industry to be “baking” AI and ML technology into its products, he said.

“What we’re starting to see is our industry partners coming in on proposals and … they’re baking in artificial intelligence and machine learning,” he said. “That’s exactly where we want to be.”

SOCOM is ramping up its pursuit of the technology, he said. “It’s a steep curve for us,” he said. “But we are climbing that curve, and in true SOF fashion we will continue to climb that curve very aggressively.”



© copyright 2017 All Rights Reserved.

A Product of HunterTech Ventures