Highlight

The emergence of electric cars has been one of the most striking aspects of the automotive industry in recent years. These vehicles have a different technology to traditional fossil fuel vehicles, and are often presented by manufacturers as "software-defined vehicles", an expression intended to illustrate the innovative technology of these vehicles. These vehicles are highly dependent on software and their problem is that they can be buggy, especially in the early versions. Take Apple, whose iPhone15 overheating problems were caused by software errors. There's no hiding it: software bugs can cause serious problems, and if the software is complex, identifying and correcting the problems can be complicated. Although demand for all-electric vehicles is increasing due to competitive superiority and incentives from some governments, several manufacturers (including GM, Volkswagen and Volvo) are evaluating and reviewing their development process. GM, for example, has postponed production of its electric trucks in the United States. Volvo has postponed deliveries of its new EX30 due to software "challenges". As the saying goes: prevention is better than cure. It's a common belief among consumers that you shouldn't buy a vehicle in the first year after its launch. It's important to give manufacturers time to correct the problems that inevitably arise in any vehicle in the early days after its launch, and with technological developments, manufacturers have perfected and automated their manufacturing processes, testing all components and their integration. When it comes to software, however, things are different: The complexity of the code, the difficulties in defining its requirements, the specificities of each vehicle, the need to recruit people with the right skills, the importance of defining appropriate software development and testing processes are some of the many problems faced by electric vehicle manufacturers. Some manufacturers eventually recognized that software development was not within their area of expertise. And so today we see several manufacturers making a huge effort in this area, recruiting specialists in software development and testing to help them understand and properly integrate software into their vehicles. The original article via PopSci can be read here.

Tesla is recalling almost 200,000 of its electric vehicles to fix a software problem that could prevent its rear-view cameras from working properly when the cars are in reverse. This recall includes some 2023 Tesla Models S, X and Y that have fully autonomous driving technology. According to a report released this week by the National Highway Traffic Safety Administration (NHTSA), the problem with the cars' rearview cameras was discovered when Tesla began receiving reports of this phenomenon at the end of December. Tesla is not aware of any accidents occurring due to the problem, and according to the NHTSA, the company has released a software update that may solve the problem, and that owners will be notified via letter, which will be sent by March 22. According to the report, since January, Tesla has received more than 80 warranty claims, possibly related to the problem. This is not the first time that the US giant has had to issue recalls, having done so with some regularity over the years. Most recently, the company recalled more than 2 million vehicles in December due to a fault in its Autopilot autonomous driving feature. The recall followed a two-year investigation by the NHTSA. The original article via The Hill can be read here.

If the prediction that AI will eliminate most jobs is close to being true, there is one profession that will certainly survive: software testing. Software testers will be the guardians of quality, ethics and behavior in AI-based systems. Let's look at five reasons why this is true, especially in the case of certified software testers. 1) The Human Presence in the AI World The first (and most crucial) point to consider is the irreplaceable human presence that software testers bring to the table. AI, despite its advances in logic and processing, lacks the human touch - the intuition, ability to perceive subtleties and ethical judgment that are characteristic of human nature. Testers act as the critical human interface, ensuring that AI systems are aligned not only with technical specifications, but also with human values and social norms. 2. A Safety Net in AI Implementation Safety in AI implementation is a primary concern and software testers are at the forefront of this battlefield. The risk of allowing AI systems to self-evaluate is similar to an untested power circuit. AI cannot be tested of itself. Human testers provide a safety net, offering an objective and critical perspective that prevents AI systems from operating in potentially harmful or unintended ways. Their role is crucial in sectors where the risks are highest, such as healthcare, automotive, aerospace, defense and public infrastructure. 3. the Quality Assurance of Certified Testers Given the critical nature of this function, it is important to distinguish between amateur testers and certified testers. Analysis shows that the involvement of certified software testers leads to a defect detection efficiency of over 99%. This contrasts sharply with the efficiency rate of 35% when amateurs, such as programmers, carry out testing. In addition, the rate of introducing new problems and correcting existing problems drops significantly with the involvement of certified testers.[1] This data is not only proof of the ability of certified testers, but a clear reminder of their indispensability for this critical quality function. 4 The Evolving Complexity of AI As AI systems become more complex, the likelihood of difficult and unpredictable problems arising increases. Software testers are essential to navigating this complexity, bringing a level of knowledge, understanding and adaptability that AI, in its current state, cannot achieve. The role of software testers is evolving from mere defect identification to architects of AI safety and reliability. 5. the ethical guardianship of AI Another vital aspect that is often overlooked is the role of software testers as ethical guardians. As AI begins to make decisions that impact human lives, the ethical implications become a critical concern. Software testers serve as a checkpoint, ensuring that AI systems comply with ethical standards and social norms. This role is becoming increasingly crucial as we deal with the moral implications of AI in everyday life. The Permanent Profession in the Future of AI: Software Tester Software testers are the humans who can never be replaced in an AI-dominated future. Their role transcends mere technical supervision because they are the guardians of safety, ethics and reliability in an automated world. As we venture deeper into AI, the need for competent, insightful, certified and ethical software testers will increase. They're not just keeping up with technological advances - they're also actively shaping a future where AI serves humanity safely and responsibly. In the narrative of our AI-driven future, software testers are not just participants, they are the ultimate professionals who ensure that this future is safe, ethical and aligned with human values. The original article via the American Software Testing Qualifications Board can be read here. ------ [1] 10 Reasons You Need Testers Certified by ASTQB - ISTQB in the U.S.

The British government is examining how to clear and compensate hundreds of post office counter managers punished for accounting errors resulting from computer glitches - "We're dealing with it and we want to put it right. The money has been put aside. What we're looking at now is how we can speed up the process," Prime Minister Rishi Sunak told reporters on Monday during a visit to Accrington, in the north of England. The Prime Minister confirmed that the Justice Minister, Alex Chalk, and the Secretary of State for the Economy, Kevin Hollinrake, will meet this afternoon to discuss a strategy. Some politicians have suggested that all wrongly convicted people should be cleared collectively through legislation to unlock compensation. More than 700 branch managers were accused of fraud, theft or falsification of accounts by the British Post Office between 1999 and 2015 when, in reality, it was a computer glitch. It was later discovered that the discrepancies in the accounts were caused by the computer system installed by the Japanese company Fujitsu, whose problems the Post Office sought to omit. For years, the state-owned company claimed that the Horizon software was reliable and forced the branch managers to reimburse the accounting deficits. The scandal was already well known, but it has been in the news again in recent days following the success of a miniseries broadcast on television by ITV, which tells the story of the campaign for justice by some of those affected. One of the law firms representing those affected has revealed that it has been contacted by at least 50 more in recent weeks. The four-part series "Alan Bates vs the Post Office" portrays the battle led over two decades by one of the Post Office's former managers to expose the truth and exonerate wronged postal workers. London's Metropolitan Police revealed on Friday that it is investigating whether the company committed fraud by taking the money obtained following these accusations and wrongful convictions. On the other hand, Fujitsu officials are also being investigated for false testimony and obstruction of justice, but in fact there have been no arrests in four years. In December 2019, a Supreme Court judge ruled that the computer system contained a series of "flaws, errors and defects" and that there was a "significant risk" of causing deficits in the accounts of postal agencies. Since then, 93 convictions have been overturned, but only 30 of these people have received full and final compensation and no one from the Post Office or other companies involved has been arrested or criminally charged. Three different financial compensation schemes have been set up over the years, resulting in the payment of more than 150 million pounds (174 million euros). A public inquiry launched in 2020 is due to be completed this year, and should reveal details about what is considered one of the biggest miscarriages of justice in the UK. The original article via Observador can be read here.

According to a recent announcement, Chinese regulators have stated that Tesla China is recalling more than 1.6 million vehicles due to problems with the steering software and door locking systems. This recall affects a number of Tesla models, including the popular S, X, 3 and Y, as well as more than 7,000 imported vehicles. On the other hand, the good news for owners of these vehicles is that the company plans to solve these problems through remote software updates, meaning there is no need for these vehicles to be assessed at a repair center. Through remote software updates, Tesla is able to simplify the recall process and provide a convenient solution for its customers. Problems with the steering software and the door locking system are major concerns for vehicle safety. The steering bug can affect the vehicle's ability to respond to the driver's commands, thus posing a risk of loss of control. Similarly, the problem with the door locking system could compromise the security of the vehicle, leaving it vulnerable to unauthorized access. Vehicle owners have been advised to stay informed and follow Tesla's instructions to initiate the necessary updates. The original article via Global Village Space can be read here.

According to a statement, Apple said it would update the iOS 17 operating system, which iPhone 15 devices run on, to fix a bug responsible for these devices overheating. The US giant is also working with developers to make changes to applications that "overload the system". The fix for this software error, which has since been released, was made via a software update sent to the affected iPhones: "This update provides important fixes; security updates and resolves an issue that may cause the iPhone to heat up more than expected," according to the update alert. Instagram, owned by Meta Platforms, has made changes to its social media app in order to prevent these overheating problems on the new iPhone with iOS 17. Other apps, such as Uber and the video game Asphalt 9, are still in the process of releasing their updates, according to Apple itself. The "iPhone 15 Pro Max heats up very easily. I'm just browsing social media and it's on fire," said one user in a post on X (formerly known as Twitter). The Wall Street Journal reported last month that the iPhone 15 Pro Max heated up to 112 degrees Fahrenheit (approximately 44 degrees Celsius) when performing processor-intensive tasks, such as gaming, and when it was being charged. It is not uncommon for new iPhones to heat up unusually during the first few days of use or when they are being restored with backup information stored in the cloud, problems that Apple has already pointed out to users. The devices can also heat up when using applications such as video games and augmented reality technology that require a lot of processing power, but the heating problems with the iPhone 15 models have gone beyond these typical situations. The original article via CBS News can be read here.

In England, an electronic sign in Norwich, where the "Wroxham" road meets the "Sprowston" road, was affected by a bug that alerted drivers to roadworks more than 160 kilometers away. The sign, which was in place for several days, has since been switched off. According to Norfolk County Council, the problem was caused by a software issue. According to the local authority, the sign should have been displaying information about works at the Heartsease traffic circle, which is less than three kilometers away, and instead informed drivers of a road closure overnight in the Leicestershire village, preventing them from reaching Melton Mowbray - "This is a software issue with our contractor, who is trying to resolve the issue," said a spokesperson for the council. The original article via the BBC can be read here.

McDonald's ice cream machines are notorious for their frequent breakdowns, and iFixit, an American repair company, is working to make them easier to repair. In a video released last Tuesday, iFixit, known for its consumer electronics disassemblies and repair guides, took apart a McDonald's ice cream machine to investigate the cause of its frequent breakdowns. According to the same video, at the time of its publication, it was estimated that at least 34% of ice cream machines were broken in New York state alone. According to the US company's analysis following this video, McDonald's ice cream machines have overly complicated error codes. This, coupled with an exclusive repair contract between McDonald's and the machine's manufacturer, Taylor, creates a heavy reliance on service technicians, making the process both costly and time-consuming. The company Kytch had already developed a device that translated error codes into simple instructions for repairing ice cream machines, but McDonald's quickly squashed this initiative, citing unproven "safety risks", according to Elizabeth Chamberlain, iFixit's repair specialist. In this context, the iFixit team decided to "take the law into their own hands" due to the limitations imposed by US copyright law. In the published video, Elizabeth Chamberlain, iFixit's director of sustainability, expressed the team's desire to create a device similar to Kytch's that could read the error codes of these machines, however, they are unable to do so due to the Digital Millennium Copyright Act (DMCA), and the software that is blocking access to the machines. Thus, iFixit and Public Knowledge have requested an exemption from the law for this purpose, similar to the one they have obtained for other products, and have therefore asked Congress to reintroduce the Freedom to Repair Act. The continuation of the original article via The Times of India can be read here.

An attack on several Norwegian government ministries has forced US federal agencies to fix an Ivanti software bug. The bug in question is a flaw related to an authentication bypass in Ivanti's Endpoint Manager Mobile (EPMM) device management software, formerly known as MobileIron Core. The US company has issued patches for this vulnerability, and the Norwegian government revealed that the flaw had been exploited in an attack that affected 12 of its ministries. In this context, the US Cybersecurity and Infrastructure Security Agency (CISA) has added this bug to its catalog of Known Exploited Vulnerabilities (KEV). In an alert issued, CISA stated that "these types of vulnerabilities are frequent attack vectors for malicious cyber actors and pose significant risks to the federal enterprise". Identified as CVE-2023-35078, the authentication bypass bug received the maximum possible CVSS rating of 10, with Ivanti warning its customers that it was essential that they patch their software immediately. According to Ivanti, this bug allowed unauthorized access to the solution's restricted functionality or resources, allowing threat actors to "potentially access users' personally identifiable information and make limited changes to the server". CISA also learned that this software bug allowed unauthenticated access to specific API paths - "An attacker with access to these API paths can access personally identifiable information (PII) such as names, phone numbers and other mobile device details for users on a vulnerable system." The Norwegian authorities did not say whether the data had been exfiltrated during the attack on their ministries, however they did say that the country's Data Protection Authority had been notified, which indicates that there may be concerns that the information has been stolen: "This vulnerability was unique and was discovered for the first time here in Norway," said Sofie Nystrøm, Director General of Norway's National Security Authority. "If we had published information about the vulnerability too early, it could have contributed to its abuse elsewhere in Norway and the rest of the world. The update is now widely available and it is prudent to announce what kind of vulnerability it is," she added. The continuation of the original article via SC Magazine can be read here.

In England, a ferryboat has suffered a breakdown caused by a software problem. The vessel, which runs between Cowes and East Cowes, on the Isle of Wight, has since been taken out of service, and according to Cowes Island Council itself, the ferry is to remain so, to be replaced by another service. This is an important means of transport for the residents of this region, as when this means of transport is inoperative, it forces a 12-mile (19 km) detour through Newport.This ferry cost more than 3 million pounds and its breakdowns and problems have been a constant since it entered service in 2017. The continuation of the original article via the BBC can be read here.