“The state-of-the-art open-access model improves on prior generative AI performance to increase developer productivity and provides developers equal access to the benefits of code generation AI, which in turn enables organizations of any size to more easily meet their full business potential”, said Harm de Vries, lead of ServiceNow’s StarCoder2 development team, and co-lead of BigCode.

StarCoder2 is trained on over 600 programming languages from The Stack v2 and some natural language code such as issues from Wikipedia, Arxiv and GitHub. The models use grouped query attention, a context window of 16,384 lexemes, with sliding window attention of 4,096 lexemes. Models 3B and 7B are trained on over 3 trillion lexemes, and model 15B on over 4 trillion lexemes.

According to the creators, the goal of StarCoder2 is to provide developers with features such as code generation, workflow generation, and text summarization, among others.

The project to develop StarCoder2 was led by ServiceNow and Hugging Face. StarCoder 2 is available in three distinct sizes: ServiceNow crafted a 3 billion-parameter model, Hugging Face developed a 7 billion-parameter model, and NVIDIA engineered a 15 billion-parameter model.

The smaller models are designed to offer powerful performance while using small amounts of compute power. According to the companies, the 3 billion-parameter model matches the performance of the 15 billion-parameter model of the original StarCoder release.

Read more:
1. Mark Zuckerberg Met with Japanese Prime Minister Fumio Kishida
2. Odysseus Mission Will Cease Operations
3. Meta Starts to Counter Disinformation

“With this release, .NET reshapes the way we build intelligent, cloud-native, applications and high-traffic services that scale on demand. Whether you’re deploying to Linux or Windows, using containers or a cloud app model of your choice, .NET 8 makes building these apps easier. It includes a set of proven libraries that are used today by the many high-scale services at Microsoft to help you with fundamental challenges around observability, resiliency, scalability, manageability, and more”, Gaurav Seth, partner director of product for developer platforms at Microsoft, wrote in a blog post.

In .NET 8, developers can integrate large language models, such as GPT, into a .NET application. To improve compatibility with generative AI workloads, the company also added several enhancements to the System.Numerics library.

The platform update includes a new code generator, called Dynamic Profile-Guided Optimization (PGO), that optimizes code based on “real world” usage and improves application performance by as much as 20%. PGO is enabled by default. Support for the AVX-512 instruction set enables parallel operations on 512-bit vectors of data to process more data in less time. Microsoft noted that it is seeing improvements in the JSON API scenario of 18%, with nearly one million requests per second with ASP.NET Core Minimal APIs.

To make it easier for developers to get started with AI, the company also created several samples and reference templates that showcase AI patterns and practices. Currently these include Customer Chatbot, Retrieval Augmented Generation, and Developing Apps using Azure AI services.

As for containers, Microsoft sought to make packaging applications easier and more secure. Every .NET image includes a non-root user for more secure containers and one-line configuration. The .NET SDK tools publish container images sans a Dockerfile and are non-root by default.

.NET 8 is supported by the Visual Studio family of tools including just-released Visual Studio 2022 17.8. Developers also can create .NET 8 applications using Visual Studio Code with the C# Dev Kit or the GitHub Codespaces template for .NET.

“We are excited to announce AI integration in multiple products to enhance developer productivity and creativity. At the same time, we’ve added one of the most-often requested features to MapForce, which is support for PDF. The new MapForce PDF Extractor will be a game changer for unlocking the volumes of data previously trapped in PDF documents and accessing it for use in data integration and ETL processes”, said Alexander Falk, president and CEO of Altova.

Altova announced the latest additions to its product line, featuring cutting-edge AI capabilities that promise to redefine the landscape of XML, JSON development, and database management. Among these innovations, the spotlight shines on the AI Assistant integrated into XMLSpy, a feature designed to significantly boost productivity.

The AI Assistant in XMLSpy streamlining XML and JSON development by effortlessly generating schemas, instance documents, and sample data through intuitive natural language prompts.

XMLSpy can also produce XSL, XPath, and XQuery code. Users have the flexibility to copy generated code, open it in a new document, or send it to the XPath/XQuery window for further tweaking and examination.

In DatabaseSpy, Altova introduces AI integration that augments the database management experience. The AI Assistant within DatabaseSpy equips users with the ability to effortlessly generate SQL statements, sample data, and table relations. Additionally, AI extensions empower users to explain, format, and complete SQL statements with unparalleled ease and efficiency.

Further enhancements include the MapForce PDF Extractor, a visual utility that simplifies the definition of a PDF document’s structure and streamlines data extraction. In StyleVision, the new Split output preview feature for XML and database report design provides users with a comprehensive view of their output in side-by-side panels, further enhancing the development process.

The Beginning – When everything finally falls into place
When Nadine Selim interviews for her dream position at Amazon, she hesitates until the last minute whether to accept it and leave her old job, which is just as good as the future offer. During this same time, Nadine is about to have one of her most exciting moments – becoming a mother. Yet she gets the job at Amazon and leaves her old position.

The tech industry and the support it provides to expectant parents
Back then, she still had no worries about being pregnant and the likelihood of losing her job when she gave birth, as she knew about the good maternity leave benefits offered by the tech giants. Benefits that many believe are partly designed to attract more members of the fairer sex into the male-dominated tech industry. And when the most long-awaited and emotional moment came for Selim, and every woman and mother-to-be – she stepped away from work for a while on maternity leave.

The email that clouds the happiness of a mother
But just a few weeks after that, Selim received an email saying she was being laid off. And so she became one of the hundreds of thousands of employees who were laid off by companies that offered good working conditions.

The tech industry and the promise it broke?
According to Slate, among the mass layoffs that have taken place in the tech sector in recent months, the number of laid-off women is quite large, and even many of them have received the unpleasant memo while on maternity leave, and Selim is one of them.

Some time ago, we told you about the story of a man, again an employee at a large tech company, who lost his job during fatherhood. The tech industry doesn’t seem to take into account the social situation of its employees, and the redundancy trend, which is slightly contained for now, threatens every single member of it, whether they are in a high position, have a lot of experience or have just become a parent and need support.

“We are deeply troubled by TikTok’s recurring pattern of providing misleading, inaccurate or false information to Congress and its users in the US, including in response to us during oversight hearings and letters,”

the senators wrote.

The letter from Senators Richard Blumenthal and Marsha Blackburn draws attention to how sensitive data about US users may be stored in China and how officials there may have access to it.

Recent reports by The New York Times and Forbes raise questions about statements made during testimony before Congress in March by Shaw Chu, TikTok’s CEO, and during an October 2021 hearing that included Michael Beckerman, TikTok’s head of public policy for the Americas. TikTok is owned by the Chinese company ByteDance.

“We are reviewing the letter. We remain confident in the accuracy of our testimony and responses to Congress.”

said Alex Haurek, a spokesperson for TikTok.

The entertainment platform has been working for years to convince the U.S. government that it can separate its U.S. operations and ring-fence U.S. users’ data amid concerns that the company could provide that information to Chinese authorities.

Stay tuned for details.

Python’s popularity has undoubtedly grown in recent years, but Julia, a relatively newer language, has recently enjoyed a lot of attention and demand. Recently, it has been increasingly argued that Julia as a programming language can compete with Python and even take its top spot. According to a new study, Python has been named the best language that developers would choose if Julia did not exist.

Python vs. Julia: Which is Faster and More Preferred?

Speed
Julia is not interpreted, therefore it is a fast programming language. It is compiled at runtime using the LLVM framework. Julia provides high speed without optimization or special profiling routines and is thus the solution to all your performance problems.

Parallelism
Both Julia and Python programming languages can execute tasks in parallel. On the other hand, Python’s techniques need data serialization and deserialization to parallelize between threads, while Julia’s parallelization is much more precise. Julia also has a less heavy parallelization syntax than Python, which lowers the barrier to using it.

Community
Every programming language should have a large and active community behind it. A programming language must have a strong following in the community. Julia has a vibrant and energetic community, but because it is a new language, the size of the community is limited.

On the other hand, Python has been around for a long time and therefore has a large community working in its favor. Julia’s developer community is still in its infancy. Python’s large community is a great advantage for developers as it provides many resources to deal with difficulties and uncertainty.

Dynamically typed
Julia and Python are dynamically typed programming languages, which means that developers don’t have to declare variables. But with Julia, you can combine and take advantage of both static and dynamic types.

Working with Shell
Julia has a really good relationship with Shell. For example, the Shell commands to check the contents of a file. Julia variables are exported to the shell as environment variables. Once opened, users can modify the file. Working with Shell commands is much easier in Julia than in Python. In this aspect, Julia is well ahead of Python.

It will put transparency, accountability and reliability at the forefront of open source AI products to counter Microsoft, OpenAI and Google. Many of the most important software products we use every day, such as Android, WordPress, PHP, Nginx and Apache, are open source.
Mozilla has pledged $30 million to develop an alternative to the closed systems developed by big tech companies that put profits first.

Mozilla’s approach, like everything else they create, will be human-first.

Mozilla is committed to helping create an Internet that hews to the following principles:
-The internet is a global public resource that must remain open and accessible
Individuals’ security and privacy on the internet are fundamental and must not be treated as optional.
-Individuals must have the ability to shape the internet and their own experiences on it.
-Free and open source software promotes the development of the internet as a public resource.
-Transparent community-based processes promote participation, accountability and trust.
-Magnifying the public benefit aspects of the internet is an important goal, worthy of time, attention and commitment.”

The company said it aims to create a decentralized artificial intelligence community to serve as a “counterbalance” to large profit-driven companies. Mozilla is focused on the benefits of AI for humanity, not on turning humans into a commodity.

The announcement states that their first two projects will focus on creating free, reliable, and open source generative AI, which is a product like ChatGPT and Bard. Their second focus is on creating a privacy-oriented recommendation system that “does not misinform or undermine our well-being.”

If it achieves this goal, Mozilla will have made an important contribution to changing what the future of AI will look like.

Freeing up focus for patient care by automating routine tasks with AI

Highly capable professionals spend an undue amount of time on routine tasks that pull their attention away from the patient. For example, in radiology departments, imaging staff say almost 25% of their work is inefficient and could be automated.

We now have the tools to automate such tasks to relieve the burden on imaging technologists and free up focus for patient care. AI-enabled technology can make it easier for technologists to plan and execute routine exams, helping them to get images first-time-right with enhanced confidence and precision.

We will see an increased use of AI-based automation in image-guided therapy, where physicians perform minimally invasive procedures on patients with heart disease or other conditions. Interventional physicians may spend up to two hours reporting on cases after a full day of treating patients. AI could help alleviate this workload by auto-logging different steps of the procedure – enabling physicians to give their full attention to providing patient care.

Improving access to care through virtual collaboration between providers

64% of global healthcare leaders say they’re currently investing heavily in virtual care. But while this healthcare technology trend is generally seen as impacting the patient-provider relationship, the rise of virtual collaboration between providers themselves offers equally promising opportunities.

The adoption of tele-ICUs in acute care is a prime example. A tele-ICU can extend critical care resources to the bedside, no matter where the hospital is.

By using a live collaboration platform, experienced sonographers can remotely assist their local counterparts with exams, while colleagues can use the same platform to discuss patients’ medical status with them.

Photo Credits: Philips

Connecting the hospital to the home with virtual care

Health technology innovations are enabling people to better engage with providers, access more care overall, and get the resources they need to stay informed while taking more preventative healthcare measures.

Healthcare providers are turning to remote patient monitoring and virtual visits in order to maximize access to care, strengthen connections with at-risk populations, minimize the risk of COVID-19 transmission, provide more convenient experiences and reduce the strain on scarce hospital resources.

According to the World Bank and World Health Organization (WHO), half the world still lacks access to quality care, and without drastic action, the WHO estimates that 5 billion people will be unable to access healthcare by 2030. World-leading healthcare organizations continue to pioneer digital health technologies that bring care closer to patients and partner with local providers to keep improving access to quality healthcare for everyone, no matter where they live.

Offering more services at facilities closer to patients

Another healthcare technology trend enabling more equitable access to care is the move from hospital-centric to locally delivered care. Patients who have traditionally had to travel long distances for state-of-the-art or specialized healthcare can access these technologies via smaller (and sometimes even mobile) healthcare facilities.

Out-of-hospital settings like local diagnostic hubs, Virtual Care Stations, and even technology-equipped container structures can offer closer-to-home healthcare that’s especially vital for people living in rural and underserved communities. Primary care facilities and local clinics, through a combination of virtual specialist support and technologies such as ultrasound and X-ray imaging, can now function as remote hospitals, providing services such as ultrasound exams to pregnant women. If a patient comes in with a broken wrist, a local practitioner can make an X-ray and immediately share it with a remote traumatologist, radiologist or emergency doctor for remote consultation instead of requiring the patient to make a lengthy trip to a big-city hospital.

Photo Credits: Philips

Certain types of specialist treatment have also become more routine, giving rise to the growing trend of office-based labs and ambulatory surgery centers that offer patients access to potentially life-saving minimally invasive procedures closer to their communities.

In the near future, one can imagine nurses routinely performing ultrasound exams at retail or community centers, using a portable handheld device with support from a remote expert sonographer and AI algorithms that can extract basic information from the scan.

Integrating data to enable clinical collaboration and patient-centered care

Multidisciplinary care teams are trying their best to connect and keep track of all relevant information pertaining to their patients – from a variety of subspecialty reports to information about patients’ current condition, demographic data, and health history. Deciding on the right therapy is equally complex, especially in areas like oncology where physicians and patients may face a bewildering number of choices.

Integrated diagnostics is a rapidly evolving field where organizations have smart, connected systems in place that give them comprehensive and actionable insights into their patient populations by uniting imaging, monitoring, laboratory, genomics and longitudinal data. Multidisciplinary tumor boards are an example of this healthcare technology trend. With integrated diagnostics solutions, care teams can now deliver comprehensive, patient-centric cancer care, uniting patient data from disparate sources in dashboards, thus facilitating collaborative diagnostics, treatment and follow-up decisions.

Using predictive analytics to proactively manage transitions of care

Using the power of AI and predictive analytics, we can now extract relevant insights on patient flow and patient care needs from vast amounts of real-time and historical hospital data, to predict capacity needs in the next 24 to 48 hours. After initial validation, the resulting algorithms can be updated on a regular basis to take recent trends and circumstances into account – offering clinical and operational teams the real-time, actionable insights they need to make timely and effective decisions.

By embedding these data-driven practices into everyday management of patient flow, healthcare systems will be able to get the most out of precious resources and manage transitions of care more effectively across the patient journey, from hospital admission to discharge and back into the home – ensuring that the patient gets the right care in the right place at the right time.

Increasing flexibility and circularity through service business models

Healthcare organizations are increasingly implementing ‘as a service’ business models. For example, many are turning to SaaS (Software as a Service), whereby cloud-based systems give physicians, staff and even patients the opportunity to access data remotely, share critical insights in real time and collaborate more effectively, for better care coordination. SaaS models generally require a lower initial investment, and give organizations the financial flexibility to ’pay as they go,’ leading to more predictable costs.

Another benefit of SaaS models is their overall energy efficiency. When companies move to the cloud, their average server utilization can approach 65% compared with just ~15% on-premise. And while the PUE (Power Usage Effectiveness) of the average data center is 1.7, the cloud industry average is just 1.2. In short, this means fewer servers are required, and that any servers that are in use are consuming less energy.

Decarbonizing healthcare through renewable energy and efficiency

The healthcare sector is a major source of greenhouse gas emissions, accounting for over 4% of global CO₂ emissions. Health leaders across the world recognize the health implications of climate change – and the industry’s responsibility to take action.

The greatest scope for improvement lies in the transition to renewable energy, as well as in improving the lifetime energy efficiency of the devices and systems used in healthcare. Electricity consumption while in use accounts for the vast majority of these products’ total environmental impact. On a global level, it is primarily the combined effect of renewable energy and energy efficiency that will drive decarbonization.

Photo Credits: Philips

Helping health consumers make sustainable choices through eco-labeling

In the rapidly expanding world of e-commerce, it is not unusual for consumers – including those buying personal health devices – to buy the same product in two or three formats to try them out at home. According to Pitney Bowes, 51% of shoppers consciously overbuy online, knowing they’ll return the items they don’t want. But when it comes to returns, too often, somewhere in the chain, it ‘pays’ to destroy the product and avoid the costs of transport, storage, relabeling, etc., rather than going back to the source.

Highlight the environmental benefits of ‘rethinking returns’ through better-informed ordering. In Europe, we see more and more brands signing up to the EU’s Green Consumption Pledge to raise awareness of products’ environmental footprint. One way they can do this is through the adoption of eco-labeling.

Turning innovations and commitments into measurable outcomes

Being able to measure and report outcomes in granular detail is central to proving the value of healthcare services, helps providers compare treatment options, and reveals opportunities for improving or even redefining care. Providing granular outcome data is increasingly becoming a requirement for unlocking reimbursement pathways, as the healthcare industry transitions away from fee-for-service models and towards value-based care.

As pressure mounts for healthcare companies to step up their actions to safeguard the health of people and the planet, there is an increased demand for proven results. This is why healthcare organizations are putting more resources into measuring not just clinical outcomes, but outcomes relating to social impact and environmental sustainability, too – such as proof of carbon-neutrality, or use of renewable energy sources.

“I would attribute the success to the quality and enormity of data.”

The Dubai Government knows AI cannot take off big sans Internet of Things (IoT) and big data, he adds.

The country is already deploying algorithms, data sharing and AI-integration in wide-ranging areas offering smart solutions and services in sectors like healthcare, education, transportation and public safety.

The Economist Intelligence Unit shared that 86% of banks and insurance companies would increase AI-related investment by 2025.

Smart Dubai Office has launched several strategies to harness AI and Blockchain. The goal is to improve government services and setting up an ‘experience lab’ run by a dedicated team that gathers citizen feedback to redesign and transform service delivery.

Drawing one’s attention to the United Nations warning that a comprehensive shift towards digitisation would require a new type of responsibility. The Telecommunications Regulatory Authority (TRA) Director General Hamad Obaid Al Mansoori says the private sector should also actively participate in the government’s efforts to give an ethical and humane dimension to digital transformation plans. He added:

“Profits should not be the sole driver or purpose.”

That’s the reason that is prompting Digital Dubai to shape global conversation around AI ethics, establishing Dubai as a thought leader in AI adoption across public and private sectors.

Sivaraman believes that continued conversation would prompt the community to eventually buy-in despite the fact Digital Dubai’s Ethical Tool Kit that supports industry, academia and individuals in understanding how AI systems can be used responsibly is not mandatory.

With the “Value of Touch”, NIVEA wants to bring attention to the power of touch, using NFTs to discuss the value society currently attributes to it, and to spark a discussion about how touch can be valuable and powerful, even if freely available to all.

The artist behind NIVEA’s NFT art piece is Clarissa Baldassarri. She is an Italian visual artist who suffered from temporary visual impairment early on in her career as a painter. The limitations she experienced inspired her to rediscover art through what she calls a conscious touch.

NIVEA’s NFT art will be minted on Polygon and will be available as a limited edition for free via a dedicated NIVEA website.