Feed aggregator

Tropical Storm Francine has formed in the Gulf of Mexico and is expected to become a hurricane before it reaches the U.S. Gulf Coast on Wednesday.

(Image credit: NOAA)

We've found the best deals around on the Apple Watch Ultra and Ultra 2, so you can enjoy a new watch for less.

You can watch the event without a cable or satellite TV subscription.

While Wyoming has relatively low electricity rates and no state incentives, there are still some federal incentives that could make installing rooftop solar in Wyoming a smart decision, depending on your situation.

Donald Trump is already promoting outlandish conspiracies ahead of Tuesday’s presidential debate with Kamala Harris.

This one common daily habit that one-third of contact wearers are guilty of doing can cause severe issues to your vision health.

You should dispute unauthorized charges, but false claims could end up costing you instead.

Fubo's current streaming deals are offering up to $30 off on multiple popular plans right now.

The U.S. placed third in the overall medal count, behind China and Great Britain. Here's a look at some of Team USA's athletic accomplishments, from winning streaks to historic firsts.

(Image credit: Ezra Shaw)

Sony has increased the DualSense controller's price by $5 at PlayStation Direct and other retailers right around the time people are expecting a potential announcement of the PS5 Pro via a Mark Cerny-hosted Technical Presentation on September 10.

While the mid-gen refresh console has yet to be announced by Sony in an official capacity, you can now expect to pay slightly more for the standard DualSense Wireless Controller. As spotted by Wario64 on X / Twitter, the PS5's official gamepad has gone up from $69.99 to $74.99.

Meanwhile, special edition colorways, including Sterling Silver, Cobalt Blue, and Volcanic Red, have jumped from $74.99 to $79.99. The price increases have been reflected at retailers like GameStop as well as the official PlayStation Direct website.

The premium DualSense Edge controller has, at the time of writing, retained its $199.99 price tag. Plus, the DualSense Charging Station also remains at $29.99. It also looks like the price has increased in the UK, too. The standard colorways have gone from £59.99 to £64.99 at PlayStation Direct, though they are still available for their original price at Amazon right now, likely while stocks last. If you're in the UK and planning to pick up a DualSense controller soon - either as a replacement or a spare - it may be best to shop now at Amazon before a potential increase here.

The price increase here will undoubtedly catch some folks off guard, especially given Sony has made no official announcement of it. The company may make a brief statement about the increase during tomorrow's PS5 Technical Presentation, though this is considered to be centered on the much-speculated PS5 Pro.

You might also like...

• Marvel vs. Capcom Fighting Collection: Arcade Classics review: the best way to experience a treasure trove of long-lost fighting games
• Astro Bot interview: how Team Asobi built a future star from three decades of PlayStation gaming history
• Astro Bot has reminded me why I fell in love with the DualSense controller

iPads Pros with an M2 chip are reliable, quick and now cost much less with this Woot deal.

Cisco’s official merch store has been the subject of a cybersecurity attack that may have resulted in compromised customer information, including payment card details.

A report by The Register claims suspected Russia-based attackers injected data-stealing JavaScript into the company’s merch store thanks to a flaw in Adobe's Magento platform.

Despite the potential severity of the issue, Cisco has confirmed no credentials were compromised during the attack, which it says was remediated swiftly.

Russian hackers target Cisco merch store

“A Cisco-branded merchandise website that's hosted and administered by a third-party supplier was temporarily taken offline while a security issue was addressed," the company noted.

The attackers exploited a vulnerability tracked as CVE-2024-34102, which affects Adobe Commerce versions 2.4.7, 2.4.6-p5, 2.4.5-p7, 2.4.4-p8 and earlier. Arbitrary code execution is possible through the vulnerability, which has been awarded a critical 9.8 severity score on the CVSS scale.

Although Adobe has issued a security patch, it’s believed as many as 75% of firms using Adobe’s tool have not applied the fix, including the Cisco merch store.

According to c/side security workers, the script was hosted on a domain associated with an IP address located in Russia. Moreover, the domain was registered just days before the attack, raising suspicions that it could have been a “fly-by-night operation designed for quick exploitation.”

While the attack may have been spotted early enough, it serves as a gentle reminder of the importance of maintaining up-to-date software and security patches in an increasingly digital world where cyberwarfare is becoming an escalating threat.

A Cisco spokesperson added: “Based on our investigation, the issue impacted only a limited number of site users, and those users have been notified.”

More from TechRadar Pro

• New Golang malware capable of cross-platform backdoor attacks spotted in the wild
• Downloaded something dodgy? Here’s the best malware removal
• Check out the best endpoint protection software

Sitting all day can be detrimental to your health. Here's how to combat its effects.

Review info

Platform reviewed: PC
Available on: PS4, Nintendo Switch, PC (Xbox version coming in 2025)
Release date: September 12, 2024 

The fact that Marvel vs. Capcom Fighting Collection: Arcade Classics exists is a miracle - a veritable glitch in the timeline. After both companies seemingly burned its bridges after Marvel vs. Capcom Infinite’s disastrous reception, any kind of comeback for the Versus series became one of gaming’s white whales.

And yet here we are in 2024, with a brand new collection that makes seven Capcom-developed Marvel games available to play once again on modern consoles and PC. Sure, Marvel vs. Capcom 2: New Age of Heroes is handily the standout title here, but the other six games (some of which have never received console ports before) are plenty worth checking out for casual and curious players.

Those expecting a content-rich experience won’t find it here, though. These are your bread-and-butter arcade ports and thus lack any additions brought to various other console ports over the years. However, with casual and ranked online play support alongside in-depth training modes and music and image galleries, there’s plenty to enjoy if you want to pick up one of these fighting games for regular play.

Reality stone

(Image credit: Capcom)

Marvel vs. Capcom Fighting Collection: Arcade Classics includes seven arcade titles, and they are as follows: 

• X-Men: Children of the Atom (1994)
• Marvel Super Heroes (1996)
• X-Men vs. Street Fighter (1996)
• Marvel Super Heroes vs. Street Fighter (1997)
• Marvel vs. Capcom: Clash of Super Heroes (1998)
• Marvel vs. Capcom 2: New Age of Heroes (2000)
• The Punisher (1993)

The collection is available on Nintendo Switch, PlayStation 4 (and on PS5 via backwards compatibility) and PC. The initial omission of an Xbox announcement was puzzling to some, but Capcom has since confirmed it’ll be coming to Xbox One (and therefore Xbox Series X and Series S via backwards compatibility) sometime in 2025.

You might have noticed that one of the games in this collection isn’t like the others. The Punisher is a two-player beat ‘em up and released at a time when Capcom was arguably the king of the subgenre, launching around the same time as the Final Fight series, Dungeons & Dragons: Tower of Doom, and the legendary Alien vs. Predator. The Punisher beat ‘em up may not be as fondly remembered as other Capcom greats, but it’s well worth a playthrough or two here, especially if you can grab a buddy for a co-op session. 

Best bit

(Image credit: Capcom)

Marvel vs. Capcom 2: New Age of Heroes is the crown jewel of this collection. This 3-vs-3 team fighter features a colossal 56-character roster shared between iconic Marvel and Capcom representation. It’s not the most balanced fighter out there with a very clear set of top-tier (and low-tier) characters, but there’s so much fun to be had in its assist system that lets you tag in fighters for massive combo potential. 

Then you have the run of fighting games from 1994’s X-Men: Children of the Atom to 1998’s Marvel vs. Capcom: Clash of Super Heroes. While Marvel vs. Capcom 2 may be the culmination of these games that came before it, I’d implore you not to write them off as novel curios here - especially as they offer plenty in terms of gorgeous visuals and compelling gameplay systems.

Marvel Super Heroes, for example, turns the iconic Infinity Gems into a gameplay mechanic. They drop into a match as collectible items and offer various buffs upon activation, such as increasing movement speed or adding projectiles to your attacks. Some characters also synergize with certain gems, offering unique abilities like Psylocke’s ninjitsu or temporary invincibility for Juggernaut.

The first Marvel vs. Capcom game is also incredibly good fun. It’s got one of the more unique ‘pre-Marvel 2’ character rosters, featuring the likes of War Machine and Venom on the Marvel side, and Jin Saotome (Cyberbots: Fullmetal Madness) and Strider Hiryu (Strider) among the Capcom picks. Presentation-wise, MvC 1 is phenomenal and showcases jaw-dropping team attacks that change depending on your chosen team of two. 

Getting jazzy

(Image credit: Capcom)

The centerpiece of the collection is unarguably Marvel vs. Capcom 2: New Age of Heroes. As one of the most well-documented and highly-requested fighting games to receive the port treatment, very little can be said about it that hasn’t been already. While it was infamously looked down on at the time of release by seasoned fighting game players, it’s since grown to become one of the most beloved games in the community for a number of reasons.

MvC 2 might be a favorite among fighting game pros, but it always had extremely strong casual appeal. It was the ultimate ‘who would win in a fight’ game years before guest characters were commonplace in the genre - and with its 3-vs-3 tag team setup and massive 56-character roster, there are countless opportunities for team synergy and dazzling combos.

Now, it is worth noting that MvC 2 is infamously, comically unbalanced. As mentioned earlier, there is a very clear set of top-tier characters that handily outclass the rest of the roster. At the highest level of play, it’s extremely common to run into teams featuring Cable, Sentinel, Magneto, or Storm largely thanks to their strong full-screen abilities. And as the game is true to its arcade version without any kind of balance changes, it’s something you’re bound to run into if you jump into online ranked play.

By no means is this a dealbreaker; I anticipate that a whole new audience of casual fans - be they modern Marvel heads or Capcom fanatics - will be mixing and matching their favorite characters on either side of the roster. Hopefully, this’ll lead to a thriving online scene for MvC 2 that hasn’t been seen since its now-delisted Xbox Live Arcade port. 

Night at the museum

(Image credit: Capcom)

Marvel vs. Capcom Fighting Collection: Arcade Classics is a fantastic way to play these seven titles on modern hardware, but it’s also a great work of game preservation in itself. Each game has its own gallery of concept art, much of which has never been disclosed by Capcom before. It’s a fascinating trip down memory lane and provides superb insight into each game’s design. Full soundtracks are also available to listen to, which is always a nice bonus.

Other welcome extras include a range of display filters, side banners (though I prefer to play with these turned off), the choice to play English or Japanese versions of each game, and the ability to toggle the selection of secret characters. Furthermore, difficulty settings and the option to set one-button specials are perfect for casual players who just want to jump in for some solo arcade mode fun.

Lastly, each game features an authentic marquee card that’s accessible from the collection’s pause menu. These will give you insight into each character’s special moves as well as how to activate unique game mechanics. Full move lists and a brand new training mode can also be accessed if you really want to get stuck in and learn your favorite characters inside and out.

Accessibility

Marvel vs. Capcom Fighting Collection: Arcade Classics is light on dedicated accessibility features, but what’s here is welcome. An option for one-button specials is a highlight, removing the need for relatively complex button inputs. Controls are also fully remappable for each title, and a variety of display options lets you choose a filter and aspect ratio of your preference. The Versus games in particular are known for bombastic on-screen visuals effects, so a setting to make the more intense effects less so could’ve been a benefit to players sensitive to visuals such as this.

Should you play Marvel vs. Capcom Fighting Collection: Arcade Classics? Play it if...

You want to know what all the fuss is about
Capcom’s Versus games are highly regarded, but it’s been impossible to play them on a readily available collection until now. If you’re a Marvel and/or Capcom fan who’s always been interested in checking out these landmark titles, this is an exceptional collection for you. 

You want to pick up a new fighting game
There are six fighting games here that are going to draw in entirely new audiences for the first time in years. If you’re interested in learning a new fighting game with classic sensibilities, then this collection comes highly recommended.

You love diving into the history of games
The collection’s various galleries are packed with concept art and development tidbits that are utterly fascinating to peruse.

Don't play it if...

You’re after something more modern and balanced
These are older games that weren’t overly concerned with balance. If the idea of incredibly cheap team compositions or broken mechanics are irksome to you, then you might want to give this collection a miss.

How I reviewed Marvel vs. Capcom Fighting Collection: Arcade Classics

I played Marvel vs. Capcom Fighting Collection: Arcade Classics on PC for around 10 hours. That included arcade mode playthroughs of all titles, a full run of The Punisher beat ‘em up game, and a good bit of time perusing gallery mode and some time in training mode, too. I primarily used the Razer Wolverine V3 Pro controller for gameplay and the Razer BlackShark V2 X to enjoy the collection’s robust audio and soundtracks.

First reviewed September 2024

Full spoilers follow for The Lord of the Rings and The Rings of Power season 2's first four episodes.

The Rings of Power's showrunners have revealed some tantalizing season 2 details about the crafting of the rings for dwarves and men.

Speaking exclusively to TechRadar ahead of The Rings of Power season 2's launch on August 29, co-creators J.D. Payne and Patrick McKay opened up on how the newly created rings will differ from their elven counterparts. And, by the sounds of it, the dwarf lords and kings of men will struggle to deal with the impact that the dark magic-infused jewelry will have on them in the Prime Video series.

Before offering some hints at far darker events – born out of said rings' formation – to come in season 2, though, Payne and McKay disclosed whether, like the elven rings, they sought any real-world inspiration for those that'll be gifted to the dwarves and men. In short: they didn't.

Celebrimbor and Mirdania (pictured) help Sauron, disguised as Annatar, to craft more rings in season 2 (Image credit: Ben Rothstein/Prime Video)

"It was always about what tone each ring should evoke, rather than [take inspiration from] any real-life historical or mythological references," Payne replied in response to my query. "For the elven rings in season one, we did take inspiration from Lalique jewelry, which has this art nouveau feeling that lent itself to the natural world, hence the vine-like and floral nature of the elven bands."

"But, for the dwarves, we wanted them to have some dwarven characteristics [with] this sort of solid, rocky, and mountainous design. Some of them have very specific detailing, such as King Durin's ring having the three peaks of Khazad-dûm, so you see the landscape reflected in the ring itself. As for the rings of men, we wanted them to be more angular and sharp, and have a very muted, unassuming color to them."

Adding to Payne's response, McKay said: "We also wanted them to feel progressively more evil with each group's creation. The dwarf rings have a slightly sinister and seductive quality to them, and then the rings for men even more so."

The elven rings weren't forged by Sauron's hand in season 1, hence their resistance to his overtures (Image credit: Amazon MGM Studios)

Sounds ominous – but why are the dwarves and men more vulnerable to Sauron's machinations via the titular rings? Despite being crafted by Celebrimbor and Sauron – at the time, the latter was masquerading as Halbrand – in The Rings of Power season 1 finale, the elven rings (and those who wear them) aren't susceptible to the influence exerted on them by Middle-earth's big bad. Indeed, while Sauron provided some insight into their creation and was later present for their forging, he didn't actually have a physical hand in making the elven rings. That's because he didn't touch any of the materials, including the mithril, that were used to craft them.

We also wanted them to feel progressively more evil

Patrick McKay

Unfortunately, the dwarves and men, who'll eventually own one of these bands, won't be so lucky. Indeed, Sauron, who spends much of one of the best Prime Video shows' second season wearing another disguise – Annatar, the so-called Lord of Gifts – was responsible for adding a new batch of mithril into the mix for the dwarven rings in season 2 episode 2. Before he drops said metal into the forge, though, he pauses for a second or two; the sorcerer imbuing it with some of his dark magic that should make it easier for him to control those who wield the rings for the seven dwarf lords.

As for the rings fashioned for the nine kings of men... well, without spoiling too much, they're created later in season 2. Considering that, in The Lord of the Rings, these nine individuals become the Nazgul, Sauron's most trusted lieutenants, due to the malevolent influence that the rings have on them, Sauron/Annatar has an even more diabolical scheme hidden up his elven sleeves for their crafting, too.

There's more to come from me on The Rings of Power season 2 front, so keep your eyes trained on TechRadar over the next four weeks as its finale draws closer. In the meantime, read more of my exclusives with its cast, including why "there has to be a cost" with the dwarven rings' creation and usage and how Sauron and Celebrimbor's relationship goes to "some dark places" in season 2's latter half.

You might also like

• Who is The Stranger in The Rings of Power? Season 2 episode 4 of the hit Prime Video show just dropped a huge hint about the wizard's real name
• The Rings of Power season 2 is a certified Prime Video hit – here are 4 more fantastically bloody fantasy shows to stream
• Get the lowdown on the best Prime Video movies worth streaming today

Apple's commerce site says it's "making updates' and to "check back soon" as we await the expected unveiling of the iPhone 16 line, the latest Apple Watch and AirPods, and updates on the Apple Intelligence rollout.

As more and more organizations embrace Artificial Intelligence (AI) and Machine Learning (ML) to optimize their operations and gain a competitive advantage, there’s growing attention on how best to keep this powerful technology secure. At the center of this is the data used to train ML models, which has a fundamental impact on how they behave and perform over time. As such, organizations need to pay close attention to what’s going into their models and be constantly vigilant for signs of anything untoward, such as data corruption.

Unfortunately, as the popularity of ML models has risen, so too has the risk of malicious backdoor attacks that see criminals use data poisoning techniques to feed ML models with compromised data, making them behave in unforeseen or harmful ways when triggered by specific commands. While such attacks can take a lot of time to execute (often requiring large amounts of poison data over many months), they can be incredibly damaging when successful. For this reason, it is something that organizations need to protect against, particularly at the foundational stage of any new ML model.

A good example of this threat landscape is the Sleepy Pickle technique. The Trail of Bits blog explains that this technique takes advantage of the pervasive and notoriously insecure Pickle file format used to package and distribute ML models. Sleepy Pickle goes beyond previous exploit techniques that target an organization's systems when they deploy ML models to instead surreptitiously compromise the ML model itself. Over time, this allows attackers to target the organization's end-users of the model, which can cause major security issues if successful.

The emergence of MLSecOps

To combat threats like these, a growing number of organizations have started to implement MLSecOps as part of their development cycles.

At its core, MLSecOps integrates security practices and considerations into the ML development and deployment process. This includes ensuring the privacy and security of data used to train and test models and protecting models already deployed from malicious attacks, along with the infrastructure they run on.

Some examples of MLSecOps activities include conducting threat modelling, implementing secure coding practices, performing security audits, incident response for ML systems and models, and ensuring transparency and explainability to prevent unintended bias in decision-making.

The core pillars of MLSecOps

What differentiates MLSecOps from other disciplines like DevOps is that it’s exclusively concerned with security issues within ML systems. With this in mind, there are five core pillars of MLSecOps, popularized by the MLSecOps community, which together form an effective risk framework:

Supply chain vulnerability

ML supply chain vulnerability can be defined as the potential for security breaches or attacks on the systems and components that make up the supply chain for ML technology. This can include issues with things like software/hardware components, communications networks, data storage and management. Unfortunately, all these vulnerabilities can be exploited by cybercriminals to access valuable information, steal sensitive data, and disrupt business operations. To mitigate these risks, organizations must implement robust security measures, which include continuously monitoring and updating their systems to stay ahead of emerging threats.

Governance, risk and compliance

Maintaining compliance with a wide range of laws and regulations like the General Data Protection Regulation (GDPR) has become an essential part of modern business, preventing far-reaching legal and financial consequences, as well as potential reputational damage. However, with the popularity of AI growing at such an exponential rate, the increasing reliance on ML models is making it harder and harder for businesses to keep track of data and ensure compliance is maintained.

MLSecOps can quickly identify altered code and components and situations where the underlying integrity and compliance of an AI framework may come into question. This helps organizations ensure compliance requirements are met, and the integrity of sensitive data is maintained.

Model provenance

Model provenance means tracking the handling of data and ML models in the pipeline. Record keeping should be secure, integrity-protected, and traceable. Access and version control of data, ML models, and pipeline parameters, logging, and monitoring are all crucial controls that MLSecOps can effectively assist with.

Trusted AI

Trusted AI is a term used to describe AI systems that are designed to be fair, unbiased, and explainable. To achieve this, Trusted AI systems need to be transparent and have the ability to explain any decisions they make in a clear and concise way. If the decision-making process by an AI system can’t be understood, then it can’t be trusted, but by making it explainable, it becomes accountable and, therefore, trustworthy.

Adversarial ML

Defending against malicious attacks on ML models is crucial. However, as discussed above, these attacks can take many forms, which makes identifying and preventing them extremely challenging. The goal of adversarial ML is to develop techniques and strategies to defend against such attacks, improving the robustness and security of machine learning models and systems along the way.

To achieve this, researchers have developed techniques that can detect and mitigate attacks in real time. Some of the most common techniques include using generative models to create synthetic training data, incorporating adversarial examples in the training process, and developing robust classifiers that can handle noisy inputs.

In a bid to quickly capitalize on the benefits offered by AI and ML, too many organizations are putting their data security at risk by not focusing on the elevated cyber threats that come with them. MLSecOps offers a powerful framework that can help ensure the right level of protection is in place while developers and software engineers become more accustomed to these emerging technologies and their associated risks. While it may not be required for a long time, it will be invaluable over the next few years, making it well worth investing in for organizations that take data security seriously.

We've featured the best online cybersecurity course.

This article was produced as part of TechRadarPro's Expert Insights channel where we feature the best and brightest minds in the technology industry today. The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/news/submit-your-story-to-techradar-pro

The explosive popularity of generative artificial intelligence is disrupting the business world as enterprises race to apply the transformative power of GenAI chatbots to supercharge their business processes.

Yet as more employees adopt new generative AI tools like ChatGPT and Copilot in their daily roles, they’re usually doing so without a second thought to the larger security implications. IT teams are challenged to monitor each new software instance with limited visibility among sprawling networks of SaaS tools. Many AI projects that employees spin up go undetected by IT, exposing their organizations to shadow IT.

The concept of shadow IT involves the use of IT systems, devices, software, and services without explicit approval from the IT department. Most shadow IT is not introduced into an organization with malicious intent. Workers are burdened with a growing list of responsibilities in an ever-accelerating business market, so many of them turn to shadow IT to get their jobs done. Shadow IT is often easier to use than internal alternatives, has less red tape, or is a better fit for their style of work.

However, many IT teams are not prepared for the risks that these programs pose to network management and data security. Consider that 90% of employees who use unsecure practices do so despite knowing that their actions will increase risks for their organizations, according to Gartner. And fully 70% of employees who use ChatGPT hide that use from their employers, according to a survey by Fishbowl.

Risky climate

In addition, 9% of workers have admitted to pasting their company data into ChatGPT, and an average company leaks confidential information to the chatbot hundreds of times each week, according to Cyberhaven. ChatGPT then incorporates all that data into its public knowledge base for sharing with other users.

In this risky climate, budgets for generative AI projects are expected to almost triple between 2023 and 2025, rising from an average of 1.5% of IT budgets to 4.3% within two years, according to survey data from Glean and ISG. Larger companies will allocate still more for AI, with 26% of firms over $5 billion in revenue budgeting more than 10% toward generative AI by 2025. And more than one-third of survey respondents (34%) said they were willing to implement generative AI quickly despite the risks of negative outcomes.

SaaS shadow IT is probably one of the biggest hidden risk factors that IT leaders face today. Most people who utilize shadow IT tend to think that they’re just using a productivity tool. However, organizations have found over and over again that there is a high risk associated with shadow IT adoption.

Detecting Shadow IT and protecting data security

Every cyber program is built around defending data, but if that data exists within shadow IT tools, then it remains unprotected. That’s why it is so important to discover what shadow IT exists in your environment, build a plan for when it happens – not if – and foster a culture that still promotes employee problem-solving while adhering to IT policy.

IT teams can apply several important considerations and precautions to maintain control over AI tools and protect their organizations from potential risks. The most effective way to detect shadow IT is on-device where the user is, as other forms of detection can miss critical information. Going to the source of shadow IT, which is the user, is the most effective approach.

After developing an inventory of shadow IT, organizations can compare the anomalies to sanctioned IT tools, survey the anomalous users, and use this information to better understand work trends, problems, and solutions. It is important to approach shadow IT users with an open mind versus shutting down the adoption. There are business problems being solved with the use of these tools, and IT teams need to understand what that need is and work collaboratively with users to ensure they have the tools they need, while keeping data secure.

Remember, shadow IT tools are only “shadowy” until they’re not. Once discovered and brought out of the shadows, the next step is to move these IT tools through procurement and internal processes for sanctioned purchases to ensure visibility and compliance.

All new AI tools should be properly managed, as shadow IT within an organization can introduce serious compliance, security, and business risks. However, recognize that shadow IT users are really just “intrapreneurs” who are seeking new solutions to existing problems. By trying to understand the reasons behind their adoption of shadow IT, organizations can identify opportunities to solve business problems that they may not currently comprehend.

Of course, you may find that some of these shadow IT tools don’t fit within the proper IT framework of rigorous controls. But once you’ve discovered the underlying user problems being solved along the way, the users and central IT can develop plans to solve these issues in a more formal and productive way.

We've featured the best online cybersecurity course.

This article was produced as part of TechRadarPro's Expert Insights channel where we feature the best and brightest minds in the technology industry today. The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/news/submit-your-story-to-techradar-pro

The Internet has become the new backbone of corporate operations. Delivery services, banking, VPN connections from anywhere and everywhere - it’s all powered by the Internet as the new delivery mechanism for customer and employee applications and services.

Guaranteeing access to an always-on experience of those digital services is essential to any business. While predictive analytics and AI-powered intelligence allow us to build forecasting models that help optimize performance and minimize downtime, outages on the Internet can permutate in an infinite number of ways. But with Internet outages occurring on external networks and within third-party providers that sit outside the owned IT perimeter, how do you attempt to identify these outages, let alone predict and mitigate them?

No longer a finite number of events

An outage will present to the end user in a standard set of ways, including slower load times or a complete inability to access an application or service. Often, there’s commonality in the underlying pattern - or the chain of events - that led to that outage occurring.

In isolation, each pattern is detectable and observable. Most IT teams will conduct a post-incident analysis to map out the pattern or sequence of events that led to an outage. This helps understand the chain reaction of events in detail such that if the same pattern was to repeat in the future, it can be detected and an intervention made before it ends in a disruption that impacts users.

The challenge facing operations teams today is that things are no longer this simple, and outages are no longer based on a finite number of isolated events.

The multi-layered and multi-interdependent outage

Networks and applications have grown in complexity and this has influenced the characteristics of outages. In particular, the underlying patterns of system behavior that cause outages aren’t as repetitively predictable as they once were. Outage causes today are a lot more intricate and harder to diagnose. For instance a system or application no longer follows a linear client-network-server architecture; instead, it operates as a “mesh” of connectivity links, IT infrastructure and software components. The challenge for ops teams is that a mesh architecture dramatically increases the number of interconnected components and therefore permutations of conditions that can cause an outage. Compared to a more linear architecture, connections between components in the mesh and the number of permutations or sequences that can form an outage pattern are both exponentially higher.

In addition, the number of components in the mesh is also ever-changing. As more features are added to an application, more components or third-party services are incorporated into the application’s end-to-end delivery chain - and into the mesh that supports it. The complexity of the application grows, and so does the range of potential causes that can bring part or all of the application down. And it’s not just the direct dependencies that are a concern; third-party infrastructure services and components come with their own interdependencies, with systems and services that are often several steps removed from view.

Is an unpredictable pattern even a pattern?

These outage patterns don’t manifest in predictable ways.

To have the best possible chance of accurately pattern-matching in this scenario, organizations need a reliable way to ‘read between the lines’ - to understand the intricate interplay of events and patterns being observed, and how that contextually relates to the performance of their specific application or infrastructure.

That level of contextual insight across any and all domains, even the ones that sit outside of enterprise visibility and control, demands a new approach to how we think about outage detection and mitigation.

Managing such a globally vast network, then, that includes networks and domains outside of enterprise control, requires a new approach to the level of data and contextual insight that IT leaders now need to care about.

When data-driven insight goes beyond enterprise scale

When it comes to seeing, but also predicting outages, it is access to high-fidelity data across all environments that matter - including cloud and the Internet - that will ultimately allow us to identify and navigate this new world of patterns on patterns on patterns, surfacing where a performance problem exists, why, and if it matters. Visibility across the end-to-end service delivery chain to see, correlate, and triangulate all patterns that matter to assuring always-on digital experiences.

So while outage patterns may permutate in perpetuity, so too are new technologies accelerating beyond human scale. Done right, new technologies will allow us to see outages within and beyond our perimeter, wherever they may occur, as well as power a new level of intelligence to generate the automated insight required to forecast all the different patterns—and the recommended action to avoid them.

We've featured the best cloud backup.

This article was produced as part of TechRadarPro's Expert Insights channel where we feature the best and brightest minds in the technology industry today. The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/news/submit-your-story-to-techradar-pro

US car rental giant Avis has confirmed it was hit by a data breach early last month affecting customer data.

In a letter to affected customers, Avis confirmed an “unauthorized third party” broke into one of its business applications, enabling the threat actor to obtain sensitive and personally identifiable information.

The attack, which is believed to have taken place between August 3 and August 6, was discovered on August 5. Avis promises to have worked with the relevant authorities and cybersecurity experts, however the extent of the attack remains unconfirmed.

Avis confirms customer data breach

Following the discovery, further research into the attack revealed on August 14 that some personally identifiable information had been accessed. Redacted information in the letter makes it impossible to know precisely which data was obtained besides customer names.

TechRadar Pro has asked Avis to confirm this, together with the estimated number of affected customers and their regions, but the company did not immediately respond.

In its warning to customers, however, the company confirmed: “We have taken steps to deploy and implement additional safeguards onto our systems, and are actively reviewing our security monitoring and controls to enhance and fortify the same.”

Consequentially, Avis will offer affected customers a one-year membership for credit monitoring via Equifax, with codes redeemable until December 31.

Avis says those affected by the breach should review and monitor their account statements and credit history for unauthorized activity.

In light the attacker may have gotten access to email addresses, phone numbers and home addresses, Avis customers together with anyone with an online presence should be wary of phishing attempts, scams and identity theft.

Other basic internet hygiene steps, like regularly updating passwords with secure alternatives and enabling two-factor authentication, are also advisable.

More from TechRadar Pro

• Here’s the best internet security suites
• We’ve rounded up a list of the best VPN with antivirus
• 170 million strong data leak traced to US data broker