With enhancements like larger context size and improved completion quality, Stable Code 3B aims to push the boundaries of AI-assisted software development.

At just three billion parameters, Stable Code 3B is designed to run efficiently on readily available hardware like laptops—unlike larger models which require expensive specialised chips. Despite its smaller size, the company claims it matches or exceeds the code completion quality of models over twice its size. 

The system builds on Stability AI’s Stable LM natural language model with additional training on software engineering data like code repositories and programmer forums. It covers 18 programming languages including Python, JavaScript, Java, C++, and Go.

The model’s training process witnessed optimisation through the incorporation of Rotary Position Embeddings (RoPE), expanding the context size for improved performance. This technique, also employed by Meta’s Llama 2 Long, allows for context lengths up to 100k tokens.

Beyond simply suggesting new lines of code, it can also fill in large missing sections in existing code. This advanced ability is known as Fill in the Middle (FIM) and allows it to automatically write entire functions or components.

The field of AI-generated code has attracted intense interest from tech giants like Microsoft, OpenAI, and Meta. Stability AI’s new system outperforms comparable models like StarCoder and establishes it as a leader in this fast-moving space:

With impressive benchmarks and increased accessibility from its efficient size, Stable Code 3B aims to bring enhanced AI code completion to a wider audience. Its arrival promises to further accelerate the integration of generative AI into software development workflows across industries.

With systems like Stable Code 3B automating rote coding tasks, developers stand to become more productive, creative, and can focus their efforts on more complex challenges.

(Photo by Joan Gamell on Unsplash)

See also: IMF: AI could boost growth but worsen inequality

Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with Digital Transformation Week and Cyber Security & Cloud Expo.

Explore other upcoming enterprise technology events and webinars powered by TechForge here.

The post Stability AI releases Stable Code 3B for enhanced coding assistance appeared first on AI News.

The recent GitLab 16.6 November release includes the beta launch of GitLab Duo Chat, a natural-language AI assistant. Additionally, the GitLab 16.7 December release sees the general availability of GitLab Duo Code Suggestions.

David DeSanto, Chief Product Officer at GitLab, said: “To realise AI’s full potential, it needs to be embedded across the software development lifecycle, allowing DevSecOps teams to benefit from boosts to security, efficiency, and collaboration.”

GitLab Duo Chat – arguably the star of the show – provides users with invaluable insights, guidance, and suggestions. Beyond code analysis, it supports planning, security issue comprehension and resolution, troubleshooting CI/CD pipeline failures, aiding in merge requests, and more.

As part of GitLab’s commitment to providing a comprehensive AI-powered experience, Duo Chat joins Code Suggestions as the primary interface into GitLab’s AI suite within its DevSecOps platform.

GitLab Duo comprises a suite of 14 AI capabilities:

• Suggested Reviewers
• Code Suggestions
• Chat
• Vulnerability Summary
• Code Explanation
• Planning Discussions Summary
• Merge Request Summary
• Merge Request Template Population
• Code Review Summary
• Test Generation
• Git Suggestions
• Root Cause Analysis
• Planning Description Generation
• Value Stream Forecasting

In response to the evolving needs of development, security, and operations teams, Code Suggestions is now generally available. This feature assists in creating and updating code, reducing cognitive load, enhancing efficiency, and accelerating secure software development.

GitLab’s commitment to privacy and transparency stands out in the AI space. According to the GitLab report, 83 percent of DevSecOps professionals consider implementing AI in their processes essential, with 95 percent prioritising privacy and intellectual property protection in AI tool selection.

The State of AI in Software Development report by GitLab reveals that developers spend just 25 percent of their time writing code. The Duo suite aims to address this by reducing toolchain sprawl—enabling 7x faster cycle times, heightened developer productivity, and reduced software spend.

Kate Holterhoff, Industry Analyst at Redmonk, commented: “The developers we speak with at RedMonk are keenly interested in the productivity and efficiency gains that code assistants promise.

“GitLab’s Duo Code Suggestions is a welcome player in this space, expanding the available options for enabling an AI-enhanced software development lifecycle.”

(Photo by Pankaj Patel on Unsplash)

See also: OpenAI battles DDoS against its API and ChatGPT services

Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with Cyber Security & Cloud Expo and Digital Transformation Week.

Explore other upcoming enterprise technology events and webinars powered by TechForge here.

The post GitLab’s new AI capabilities empower DevSecOps appeared first on AI News.

The report, titled ‘The State of AI in Software Development,’ presents insights from over 1,000 global senior technology executives, developers, and security and operations professionals.

The report reveals a complex relationship between enthusiasm for AI adoption and concerns about data privacy, intellectual property, and security.

“Enterprises are seeking out platforms that allow them to harness the power of AI while addressing potential privacy and security risks,” said Alexander Johnston, Research Analyst in the Data, AI & Analytics Channel at 451 Research, a part of S&P Global Market Intelligence.

While 83 percent of the survey’s respondents view AI implementation as essential to stay competitive, a significant 79 percent expressed worries about AI tools accessing sensitive information and intellectual property.

Impact on developer productivity

AI is perceived as a boon for developer productivity, with 51 percent of all respondents citing it as a key benefit of AI implementation. However, security professionals are apprehensive that AI-generated code might lead to an increase in security vulnerabilities, potentially creating more work for them.

Only seven percent of developers’ time is currently spent identifying and mitigating security vulnerabilities, compared to 11 percent allocated to testing code. This raises questions about the widening gap between developers and security professionals in the AI era.

Privacy and intellectual property concerns

The survey underscores the paramount importance of data privacy and intellectual property protection when selecting AI tools. 95 percent of senior technology executives prioritise these aspects when choosing AI solutions.

Moreover, 32 percent of respondents admitted to being “very” or “extremely” concerned about introducing AI into the software development lifecycle. Within this group, 39 percent cited worries about AI-generated code introducing security vulnerabilities, and 48 percent expressed concerns that AI-generated code may not receive the same copyright protection as code produced by humans.

AI skills gap

Despite optimism about AI’s potential, the report identifies a disconnect between organisations’ provision of AI training resources and practitioners’ satisfaction with them. 

While 75 percent of respondents stated that their organisations offer training and resources for using AI, an equivalent proportion expressed the need to seek resources independently—suggesting that the available training may be insufficient.

A striking 81 percent of respondents said they require more training to effectively utilise AI in their daily work. Furthermore, 65 percent of those planning to use AI for software development indicated that their organsations plan to hire new talent to manage AI implementation.

David DeSanto, Chief Product Officer at GitLab, said:

“According to the GitLab Global DevSecOps Report, only 25 percent of developers’ time is spent on code generation, but the data shows AI can boost productivity and collaboration in nearly 60 percent of developers’ day-to-day work.

To realise AI’s full potential, it needs to be embedded across the software development lifecycle, allowing everyone involved in delivering secure software – not just developers – to benefit from the efficiency boost.” 

While AI holds immense promise for the software development industry, GitLab’s report makes it clear that addressing cybersecurity and privacy concerns, bridging the skills gap, and fostering collaboration between developers and security professionals are pivotal to successful AI adoption.

(Photo by Luca Bravo on Unsplash)

See also: UK government outlines AI Safety Summit plans

Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with Digital Transformation Week.

Explore other upcoming enterprise technology events and webinars powered by TechForge here.

The post GitLab: Developers view AI as ‘essential’ despite concerns appeared first on AI News.

With the ability to automate and streamline the testing process, enhance accuracy, and reduce costs and time spent on it, artificial intelligence (AI) has the potential to revolutionize how organizations test digital experiences in the coming years. Here are a few important ways AI can transform testing:

Automating the testing process

AI has transformed how businesses test their digital experiences. The application of machine learning algorithms and predictive analytics enables AI-powered testing solutions to simulate user behavior, develop test cases, and execute tests automatically. This automation assists organizations in saving time and costs, reducing errors, and increasing the accuracy of their testing operations.

AI-powered testing systems can automatically build test cases based on established rules or by studying user behavior, guaranteeing that the digital experience fulfils the needs and expectations of customers.

Another area where AI-powered testing solutions shine is test execution. By modeling user behavior, engaging with the digital product, and reporting results, AI can automate the test execution process. It can automatically detect problems, track flaws, and generate reports.

AI can also automate regression testing, which entails testing a digital product after modifications are made to ensure that the changes did not bring new faults. AI can detect portions of a digital product that require regression testing, build test cases, and automatically execute tests.

Another area where AI can aid automated testing is performance testing. AI can automatically simulate user behavior, generate load, and monitor system performance, finding performance issues and bottlenecks.

Finally, AI can enable continuous testing to ensure that it satisfies the appropriate quality, functionality, and user experience criteria.

Optimizing the testing process

Test prioritization is one of the most critical ways that AI may improve the testing process. AI can assess testing data to select tests based on their importance and chance of detecting flaws. This allows organizations to concentrate their testing efforts on the most critical areas, saving time and resources.

Test optimization is another method AI can be used to improve the testing process. It can examine testing data to discover redundant exams that can be deleted to enhance efficiency.

AI can automate test environment creation and configuration, ensuring that the proper environment is available at the right time. Furthermore, AI can develop synthetic test data, automate test data creation and maintenance, and assure data privacy and security.

Finally, artificial intelligence (AI) may evaluate testing results to find patterns and trends, deliver insights like areas where testing needs to be improved, recommend new tests to be added to the testing suite, and suggest process changes.

Improving accuracy

AI can increase test accuracy in a variety of ways. One way is through its ability to swiftly and accurately evaluate large amounts of data. AI can spot patterns, trends, and potential flaws in testing data that people may miss. This helps to ensure that all potential problems are recognized, lowering the chance of releasing a product with unknown flaws.

AI can also help increase testing accuracy by automating the testing process. This improves testing accuracy by ensuring that all tests are conducted consistently and accurately, lowering the possibility of unforeseen faults. AI can decrease the danger of human error and reduce the time and effort required by automating testing.

Reducing time and costs

AI can drastically cut the time and expense of testing in various ways. By automating repetitive and time-consuming processes such as test case generation, execution, and defect identification, AI can free up testers to focus on more complicated duties. This can improve the testing process’s efficiency and minimize the time and expense associated with testing.

Conclusion

Businesses can benefit from AI by testing digital interactions more efficiently, precisely, and affordably. Businesses can then provide high-quality digital experiences that satisfy customers’ expectations and demands. Additionally, businesses can gain a competitive advantage in the rapidly expanding digital market by using AI-powered testing tools.

The post How AI can transform the way enterprises test digital experiences appeared first on AI News.

The company selected 10 contests that were hosted on Codeforces – a programming competition platform with thousands of participants – to evaluate the performance of AlphaCode.

Below is an example of a problem #AlphaCode can successfully solve, using the exact information seen on @codeforces, & the program that #AlphaCode writes.

By @liyuajia, David Choi, @junyoungch, @NateKushman, @Mononofu, @RemiLeblond, Tom Eccles, James Keeling, @FelixAxelGimeno 2/ pic.twitter.com/TJu0Va46OS

— DeepMind (@DeepMind) February 2, 2022

Following the simulations, AlphaCode ranked in the top 54 percent of competitors. That means it wasn’t yet able to beat leading human programmers but could rival the average.

Mike Mirzayanov, Founder of Codeforces, said:

“I can safely say the results of AlphaCode exceeded my expectations. I was sceptical because even in simple competitive problems it is often required not only to implement the algorithm, but also – and this is the most difficult part – to invent it.

AlphaCode managed to perform at the level of a promising new competitor. I can’t wait to see what lies ahead!”

AlphaCode uses transformer-based language models to generate code “at an unprecedented scale”. A preprint paper detailing AlphaCode is available here (PDF).

Petr Mitrichev, Software Engineer at Google, commented:

“Solving competitive programming problems is a really hard thing to do, requiring both good coding skills and problem-solving creativity in humans.

I was very impressed that AlphaCode could make progress in this area and excited to see how the model uses its statement understanding to produce code and guide its random exploration to create solutions.”

DeepMind has released its dataset of competitive programming problems and solutions on GitHub to help others build on their results.

Ian Funnell, Manager of Developer Relations at Matillion, said:

“Advancements like AlphaCode are welcomed and represent huge progress in designing algorithms more effectively. AI coding in general empowers developers to pursue innovation and creativity in setting the parameters and goals, leaving the AI to actually execute them.

Ultimately, this is a catalyst for innovation—helping humans rather than replacing them. Developers are extremely capable individuals, and businesses will continue to count on them to reap valuable insights from their data to differentiate and compete.”

DeepMind has set up an interactive site to view some of AlphaCode’s solutions and dive into the model at alphacode.deepmind.com.

(Image Credit: DeepMind)

Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo. The next events in the series will be held in Santa Clara on 11-12 May 2022, Amsterdam on 20-21 September 2022, and London on 1-2 December 2022.

Explore other upcoming enterprise technology events and webinars powered by TechForge here.

The post DeepMind AlphaCode rivals the abilities of human programmers appeared first on AI News.

For AI enthusiasts, speculating about how artificial intelligence can improve software development is exciting. Will AI help create prototypes in days, instead of months or years? Will it teach human developers how to code better? AI research is broad, and the flexibility of computer programming is essentially limitless, so it’s hard to imagine what software development will look like when intelligent programs can help us interact with code.

But what many developers and tech managers don’t realise is that AI’s usefulness for development teams has made huge leaps in just the last several years. In fact, the early stages of AI-assisted software development are already here.

Automation Is Incomplete Without AI Assistance

It’s hard to find an organisation that doesn’t have efficient, agile software development as a goal, and automation technology has made agility achievable at scale. Automatically running tests has improved software quality in the last decade by allowing developers to get immediate feedback on their code changes and adjust accordingly. Automated software pipelines make use of robot assistants that generate pull requests, enabling the continuous delivery of updates. 

But the companies that have embraced the technology are sometimes finding that automation alone isn’t enough. Automated processes still have bottlenecks, most of which surround the creation of new code. For example, automating the execution of hundreds or thousands of unit tests can be done quickly, but it takes hours or weeks for the development team to write the tests themselves. Without tests to validate commits, automated pipelines promote junk. What would otherwise be an automatic process is broken up by the need for ongoing manual effort as new code (and new tests) are added. 

AI for Code

Currently, existing AI for code technology can address both of these issues in one go, by automatically writing test code that validates the rest of the automation pipeline. This type of task previously required a developer’s time, preventing them from doing the more fulfilling and value-adding work, such as creating new features. AI used for unit testing opens the door to the more complete automation of important, but slow and tedious, processes.

As might be expected, tests created by AI look different from those written by humans, but they will be produced in a fraction of the time and they function just as well, if not better, to ensure that any code-breaking changes are caught while it’s still easy to fix them. After all, as Martin Fowler has summed up in his 2006 article on Continuous Integration: “Imperfect tests, run frequently, are much better than perfect tests that are never written at all.”

In this way, AI can start to break down the trade-off between time, cost and quality of work that developers and IT managers struggle with. AI-assisted development can empower developers to create new products faster and more cost-effectively without compromising on quality. With repetitive tasks completed reliably and at speed, developers can get back to the creative tasks that attracted them to their jobs in the first place.

The Incredible Efficiency of AI 

In some industries that place a high value on the quality of their code, such as finance, AI-assisted software development is already in use. Goldman Sachs, for example, recently used AI for code to improve the efficiency of their software development. By leveraging an AI tool to write over three thousand unit tests for one legacy application with fifteen-thousand lines of code, they created an entire test suite in hours. Compared to spending an average of 30 minutes writing each unit test manually, the AI tool was able to write tests more than 180 times faster. All told, the bank managed to save over a year of developer time.

What’s Next?

As AI technology continues to advance and solutions for more use cases are developed, investments in AI for software development will become more common across industries. It won’t be long before integrating a new level of efficiency-boosting tools into the development process is a requirement in order to stay competitive and grow at scale. But in the meantime, the first iterations of AI-assisted software development are already here and giving a glimpse of what we can expect from the future of coding.

Interested in hearing industry leaders discuss subjects like this? Attend the co-located 5G Expo, IoT Tech Expo, Blockchain Expo, AI & Big Data Expo, and Cyber Security & Cloud Expo World Series with upcoming events in Silicon Valley, London, and Amsterdam.

The post Opinion: AI for software development is already here appeared first on AI News.