Over a dozen companies have suffered data theft attacks after a SaaS integration provider was breached and authentication tokens stolen.

While numerous cloud storage and SaaS vendors were targeted using the stolen tokens, BleepingComputer has learned that the majority of the data theft attacks targeted the cloud-based data warehouse platform Snowflake.

Snowflake confirmed “unusual activity” to BleepingComputer, stating that a small number of its customers were impacted.

“We recently detected unusual activity within a small number of Snowflake customer accounts linked to a specific third-party integration,” Snowflake told BleepingComputer.

“We immediately launched an investigation and, out of an abundance of caution, locked down potentially impacted customer accounts. We also notified potentially impacted customers and provided precautionary guidance to help them further protect their accounts.”

Snowflake stressed that the attacks did not involve any vulnerability or compromise of its systems.

As part of these attacks, the threat actor allegedly attempted to use the stolen authentication tokens to steal data from Salesforce, but was detected before they could succeed.

Data theft after alleged Anodot breach

While Snowflake would not confirm which third-party integration partner was linked to these attacks, BleepingComputer was told by numerous sources that the attacks stem from a security incident at data anomaly detection company Anodot.

Anodot is an AI-based analytics company that provides real-time anomaly detection for business and operational data, helping organizations automatically spot unusual changes in revenue, transactions, and system performance using machine learning. Data analytics company Glassbox acquired the company in November 2025.

BleepingComputer was told that numerous companies are now being extorted by the ShinyHunters extortion gang, which is demanding ransom payments to prevent the release of stolen data.

After learning of the attacks, the ShinyHunters group confirmed to BleepingComputer that they were behind them, claiming to have stolen data from dozens of companies this past Friday. They also confirmed their attempts to steal data from Salesforce, but said they were blocked by AI detection.

The blocked attempt comes amid a wave of data theft attacks over the past year targeting Salesforce customers.

The threat actors also claimed the attack stems from a security incident at Anodot, hinting that they allegedly had access to the company for some time.

The threat actor shared some of the companies allegedly affected by the incident, but BleepingComputer will not name them without confirmation.

Only one company, Payoneer, replied to our emails, stating that it aware of the integrator breach but was not impacted.

“We’re aware of a security incident involving a third-party service provider, Anodot. Based on our review, Payoneer has not been impacted,” Payoneer said in a statement to BleepingComputer.

Google’s Threat Intelligence Group, which has been tracking many of this year’s data theft campaigns, also confirmed to BleepingComputer that it is aware of the incident and is tracking it, but had nothing further to share at this time.

BleepingComputer has sent multiple emails to Anodot and its parent company, Glassbox, but has not yet received a reply.

Automated pentesting proves the path exists. BAS proves whether your controls stop it. Most teams run one without the other.

This whitepaper maps six validation surfaces, shows where coverage ends, and provides practitioners with three diagnostic questions for any tool evaluation.

Get Your Copy Now

U.S. victims lost nearly $21 billion to cyber-enabled crimes last year, driven primarily by investment scams, business email compromise, tech support fraud, and data breaches, the Federal Bureau of Investigation says.

The figure continues the year-over-year record trend as it is up 26% compared to 2024, when Americans lost $16.6 billion to cybercrime.

A similar uptick was recorded in the number of complaints the Internet Crime Complaint Center (IC3) received, which surpassed 1 million last year, up from 859,000 the year before.

The most frequent complaints received last year referred to phishing attacks (191,000), extortion (89,000), and investment scams (72,000), which continued to drive massive losses.

Although smaller in absolute numbers, there were still a significant number of reports for serious attack types such as business email compromise (24,700 cases), data breaches (3,900), ransomware attacks (3,600), and SIM swapping (971).

Investment fraud accounted for 49% of all scam-related incidents recorded last year and resulted in losses of $8.6 billion. However, cybercrime targeting cryptocurrency caused the largest loss, exceeding $11 billion across 181,565 cases.

Cyber-enabled fraud was present in 453,000 complaints and accounted for $17.7 billion of the total losses submitted to the IC3 in 2025.

According to the IC3, Americans over the age of 60 were hit the hardest, with reported losses of $7.7 billion, a 37% increase compared to the previous year.

For the first time, the FBI’s report includes AI-related scams, which accounted for 22,300 complaints and $893 million in losses. These schemes involved voice cloning, fake profiles, forged documents, and deepfake videos.

In two cases, attacks targeting critical infrastructure (dams and nuclear facilities), the FBI labeled the incidents as data breaches.

The most targeted critical infrastructure sectors in 2025 were healthcare, manufacturing, financial services, information technology, and government facilities.

FBI fighting back

The FBI says that it has upgraded its efforts to block attacks, notify victims, and freeze stolen funds, in some cases even being able to retrieve them.

The agency initiated 3,900 Financial Fraud Kill Chain (FFKC) interventions in 2025, successfully blocking a portion of fraudulent transactions. Of the $1.16 billion targeted by attackers, the FBI froze $679 million.

Additional efforts from the agency to prevent cyber-enabled crimes included ‘Operation Level Up’ at the start of the year, a proactive approach to prevent financial losses by identifying and alerting victims of cryptocurrency investment fraud.

Of the 3,780 victims notified last year, 78% were unaware that they were being scammed.

The FBI recommends people not to rush when they receive urgent requests and face pressure tactics, and to use all available means to verify the authenticity of the communication before sending money or data.

Those who suspect compromise by hackers or scammers are urged to report the incidents with full details to ic3.gov.

Automated pentesting proves the path exists. BAS proves whether your controls stop it. Most teams run one without the other.

This whitepaper maps six validation surfaces, shows where coverage ends, and provides practitioners with three diagnostic questions for any tool evaluation.

Advertisement

Get Your Copy Now

From being merely an auxiliary element, artificial intelligence has now become an integral part of the driving force behind businesses. Be it the analysis of large data sets or the execution of repetitive functions, the importance of artificial intelligence has already been demonstrated in multiple industries. However, the introduction of generative artificial intelligence has now added a new dimension to the capabilities that are being offered.

While traditional AI has been widely used for years, the rise of generative AI is making it important for businesses to understand how the two differ. Although these technologies are part of the same broad category, namely ‘artificial intelligence solutions,’ but they have very distinct functionalities and differences.

Understanding Traditional AI

Generative AI employs transformer models to process the context and generate outputs similar to humans.

Why Generative AI Is Driving New Opportunities

The increased interest in generative AI services is attributed to the potential they have to enable innovation and efficiency in multiple functions. 

Key Benefits of Generative AI 

1. Scalable Content Creation 

Generative AI enables businesses to generate large quantities of content within a short time, such as marketing content, reports, and product descriptions, thus helping them save time and be consistent in the content they generate.

2. Enhanced Customer Engagement 

Businesses can use AI-based chat tools to produce more personalized & engaging content, thus giving the customer a superior experience and effectively fulfilling their needs. 

3. Quicker Product Development 

Generative AI enables faster product development through the generation of prototypes, codes, and testing.

4. Personalization  

Businesses can use generative AI to create a personalized experience based on individual needs, hence providing a better experience and satisfying users.

5. Data Augmentation  

Artificial data helps improve models, especially when there is a lack of data, hence providing accurate results and improved performance.

These advantages make generative AI development a vital part of digital strategy for many organizations.

Real-World Applications Across Industries

Traditional AI Applications

• Financial fraud detection

• Predictive maintenance in manufacturing

• Inventory management

• Customer segmentation

Generative AI Applications

• AI chatbots and conversational agents

• Marketing content creation

• Code generation and debugging

• Image and video generation

• Virtual assistants and knowledge systems

These examples show how generative AI solutions extend beyond traditional automation into areas that require creativity and adaptability.

Combining Both Approaches

In fact, most organizations use a mix of traditional and generative AI to get the best out of the systems.

Example Use Case

• Traditional AI systems process customer information to find patterns

• Generative AI systems use the patterns to create personalized content

This enables businesses to use the advantages of both systems to create a more robust artificial intelligence system.

Challenges and Considerations

Traditional AI Limitations

• Limited flexibility

• Difficulty in dealing with unstructured data

• Need for manual updating for new applications

Generative AI Challenges

• Higher computational costs

• Chances of inaccurate or biased outcomes

• Need for effective governance and compliance

Understanding regulatory requirements is equally important, as generative AI regulatory compliance helps businesses manage risks effectively while adopting new technologies.

Business Impact of Generative AI

The rise of generative AI services is influencing how businesses operate and compete.

Key Areas of Impact

• Marketing and Content Creation

Faster production of high-quality content

• Customer Support

Improved interaction through intelligent chat systems

• Product Innovation

Rapid prototyping and idea generation

• Operational Efficiency

Automation of complex workflows

Organizations investing in generative ai development are finding new ways to improve productivity and deliver value.

Choosing the Right Approach

Selecting the right AI approach depends on the nature of your business needs.

Use Traditional AI When:

• Working with structured data

• Focusing on prediction and analysis

• Managing risk and compliance tasks

Use Generative AI When:

• Creating content or designs

• Building conversational interfaces

• Driving innovation and personalization

Key Factors to Consider

• Data availability and quality

• Infrastructure requirements

• Integration with existing systems

• Long-term scalability

A thoughtful evaluation helps businesses select the most suitable generative AI solution or combination of tools.

Concluding Thoughts

The difference between traditional AI and generative AI is based on how they approach problems and how they deliver the results. While traditional AI is based on analyzing data and making predictions, generative AI creates new information and allows for more interactive experiences.

As the business world continues to look into more sophisticated technology, the development of generative AI is a significant part of the current business plans. This is because it opens up more opportunities for creativity and efficiency in the customer world. 

However, traditional AI is still a viable tool in the world of data-based tasks. The most viable approach to artificial intelligence solutions is a mix of traditional and generative AI. This ensures a balance between traditional and generative artificial intelligence solutions and allows businesses to thrive and become more competitive in a more data-based world. 

As US President Donald Trump threatens wholesale demolition of Iran’s infrastructure in the midst of an escalating war, Iran now appears to have already reciprocated with its own form of infrastructure sabotage: A hacking campaign hitting industrial control systems across the United States, including energy and water utilities, that US agencies say has had disruptive and costly effects.

In a joint advisory published Tuesday, a group of US agencies including the FBI, the National Security Agency, the Department of Energy, and the Cybersecurity and Infrastructure Security Agency warned that a group of hackers affiliated with the Iranian government has targeted industrial control devices used in a series of critical infrastructure targets including in the energy sector, water and wastewater utilities, and unspecified “government facilities.” According to the agencies, the hackers have targeted programmable logic controllers (PLCs)—a type of device designed to allow digital control of physical machinery—in those facilities, including those sold by industrial tech firm Rockwell Automation, with the apparent intention of sabotaging their systems.

By compromising those PLCs, the advisory warns, the hackers sought to change information on the displays of industrial control systems, which can in some scenarios cause system downtime, damage, or even dangerous conditions. “In a few cases, this activity has resulted in operational disruption and financial loss,” it reads.

When WIRED reached out to Rockwell Automation, a company spokesperson responded in a statement that it “takes seriously the security of its products and solutions and has been closely coordinating with government agencies in connection with” Tuesday’s advisory, and pointed to documents it has published for customers on how to better secure their PLCs.

Though the advisory doesn’t specify a particular group responsible for the hacking campaign, it notes that the attacks are similar to those carried out in by the Iran-linked group known as CyberAv3ngers, or the Shahid Kaveh Group, starting in late 2023. That team of hackers, believed to work in the service of the Iranian Revolutionary Guard Corps, inflicted several waves of attacks against Israeli and US targets in recent years, including gaining access to more than a hundred devices sold by industrial control system technology firm Unitronics and most commonly used in water and wastewater utilities.

This is a developing story, please check back for updates.

After years of rumors and leaks, Apple’s iPhone Fold is finally inching closer to a launch, and according to Bloomberg’s Mark Gurman, arguably the most reliable Apple oracle, the foldable is on track for a September 2026 debut. 

Earlier this week, Nikkei Asia raised eyebrows by flagging development snags with the iPhone Fold, particularly during testing. Gurman, however, pushes back, reporting that despite the device’s complex design, Apple remains committed to its September launch window. 

Is the iPhone Fold’s launch on track?

The supply at launch might be tighter than for other iPhones, but the launch timeline itself stands. Does that mean that the purported iPhone Fold will be available to purchase alongside the iPhone 18 Pro models in September 2026? It doesn’t seem so. 

While Apple intends to reveal or showcase the iPhone Fold alongside the iPhone 18 Pro and the iPhone 18 Pro Max in September this year, shipping may follow later. In his newsletter, Gurman drew a parallel between the rumored foldable and the iPhone X. 

When will Apple chip the iPhone Fold?

Apple announced the revolutionary iPhone X in September 2007, but deliveries began in November. The Fold, at least for now, could follow a similar path, potentially hitting hands as late as December this year. 

In exchange for the long wait, buyers could get a book-style foldable that unfolds from a 5.5-inch outer screen to a 7.8-inch inner screen that mimics the aspect ratio and look of the iPad mini, has a 4.5-4.8 mm side profile, runs on a new Apple chip, and has a dual-rear-facing camera array. 

Pricing, however, is the most sensitive aspect of the iPhone Fold, which, in my opinion, can make or break the company’s product lineup. For now, the general consensus on the internet points toward a starting price of over $2,000 for the baseline variant.

The four astronauts aboard the Integrity spacecraft now headed home from their historic arc around the moon really are like the rest of us: Sometimes they reach for their smartphones to snap photos.

For the Artemis II mission, iPhone 17 Pro Max phones have been used to capture photos inside the capsule of the astronauts pondering the views of Earth and working on mission objectives. (Technically, NASA refers to them as PCDs – personal computing devices.)

Smartphones were cleared for use in space for the first time in February. In a post on X, NASA Administrator Jared Isaacman wrote, “We are giving our crews the tools to capture special moments for their families and share inspiring images and video with the world.”

Early in the mission, Commander Reid Wiseman snapped a pair of photos looking out the window with Earth behind him. Mission specialist Christina Koch and her dynamic curls in zero-gravity also captured a pensive view looking out over the planet. All three were made using the front camera — because wouldn’t you want to grab a selfie if you were in space?

The iPhone 17 Pro’s rear cameras are pulling their own weight during the mission, too. During the live broadcast as the crew approached the moon, Wiseman took a photo of the moon’s surface using the iPhone’s telephoto camera at 8x zoom. He turned the screen toward one of the video cameras mounted inside the spacecraft, creating an image of the moon’s surface alone against the darkness of the unlit cabin, with the iPhone’s signature rounded edges and Dynamic Island cutout at the top.

Advertisement

The main photo workhorses on this trip are a Nikon D5 DSLR and a Nikon Z9. The D5 is a model that has been used on several space excursions, and the Z9 is onboard as an experimental camera.

For NASA missions, every piece of equipment must be tested and certified, which is why the previously-approved D5 has a secure spot. Cameras must be resistent to space environmental factors like radiation, and safe if they’re floating around the capsule. However, the iPhones in space now are off-the-shelf models, according to a report by Jackie Watties of CNN.

The moon flyby was especially photo-intensive, with astronauts switching places several times so that two were always at windows with cameras and relating what they could see with their eyes. This photo of mission specialist and Canadian Space Agency astronaut Jeremy Hansen taking images using one of the Nikon cameras shows how some windows have camera shrouds attached. The shroud ensures that light from the interior isn’t reflected in the glass.

In a particularly relatable photo, Hansen is also using the front-facing camera of a white iPhone 17 Pro — as a portable mirror while he shaves. As the (modified) saying goes, the best selfie screen is the one you have with you.

Advertisement

The iPhone 17 Pro isn’t the first Apple product to go into space. Crew members have taken iPods, iPads and AirPods on missions since the Space Shuttle era. The Mac Portable even went up on a shuttle (and revealed that its trackball in zero-G isn’t the best option).

An Apple representative didn’t immediately respond to a request for comment.