Anthropic recently shared that its new AI model, Claude Opus 4.6, found over 500 serious security flaws in open-source software.

These flaws, known as zero-days, were found in code that people have used for years and thought was safe.

The news is a big deal because it shows that computers are now much better at finding mistakes than humans.

A new way to find bugs

Usually, finding deep security holes takes a human expert many weeks of hard work.

This new AI model works differently because it can read and think about code like a person would.

It found bugs in projects that had already been checked by other tools for millions of hours without any results.

The race between good and bad

The problem is that if an AI can find these flaws in minutes, bad people can use AI to do the same thing.

While it used to take weeks to fix a bug once it was found, a hacker can now use an exploit in just a few hours.

This creates a dangerous gap where the people attacking are moving much faster than the people defending.

Focusing on trust and people

Since systems are becoming harder to break, hackers are now trying to trick people instead.

They use AI to learn how a company works, who talks to whom, and how they approve payments.

This means the next big threats might not look like a broken computer but like a fake message from a boss.

Security teams must now move faster than ever to patch holes before the wrong people find them first.

Hackers are moving away from building their own tools and are instead using the same cloud services your company uses every day.

This change makes it very hard for security teams to spot bad behavior because it looks just like normal office work.

Because services like AWS and Google are trusted by default, many systems do not even check to see if the traffic is safe.

Hiding in Plain Sight

Attackers are now using apps like Google Sheets or Slack to send commands to their viruses.

By doing this, they avoid being caught by lists of "bad" websites since these famous platforms are almost never blocked.

They also use things like OneDrive or SharePoint to store stolen files, making it look like a regular backup.

Abusing Cloud Features

Some hackers use cloud "snapshots" to steal a whole copy of a computer's hard drive in seconds.

Instead of trying to break in through the front door, they simply share a copy of the server with their own account.

Others use small, fast tools called Lambda functions to launch thousands of tiny attacks at once, which can overwhelm traditional defenses.

Stealing Trust

Bad actors are also making fake login pages that are hosted on real cloud servers to trick workers.

When a person sees a link that starts with a trusted name, they are much more likely to type in their password.

They even use tools meant for remote work to sneak past firewalls that usually keep the bad guys out.

The cloud is no longer just a place to store files but is now a major tool that hackers use to run their entire operation.

Salesforce is warning companies to check their settings because hackers are finding ways to steal private data from public web pages.

A group of cyber criminals says they have already taken information from hundreds of websites and many large, well-known firms.

This problem happens because of simple mistakes in how companies set up their online portals for guests.

The Risk of Guest Access

Many firms use a tool called Experience Cloud to build sites for their partners and customers.

These sites have a special "guest" setting that lets people see basic info without needing a password or a login.

If these settings are too loose, hackers can use automated tools to ask for much more data than they are allowed to see.

How Hackers Steal Data

Bad actors are using a modified computer tool to scan thousands of sites at once to find these weak spots.

Instead of trying to break through strong locks, they simply look for digital doors that were left wide open by mistake.

Once they find a gap, they can pull out sensitive customer records and use them to demand money from the company.

Ways to Stay Safe

Security experts say companies must look at their guest rules right away and limit what outsiders can see.

It is important to turn off access to certain parts of the system that guests do not really need to use.

By following the rule of giving only the smallest amount of access possible, businesses can keep their most valuable data out of the wrong hands.

Keeping your digital doors locked is much easier than dealing with the mess after a hacker gets inside.

Being a security leader is changing quickly because of how smart computers are becoming.

Chief Security Officers must now learn how to use artificial intelligence to defend their companies from new kinds of attacks.

Since hackers are also using these smart tools, the battle has become a test of who has the best team and the best technology.

The Power of Humans and AI

Security experts say that AI by itself is not strong enough to win every fight.

The best defense happens when a person with business knowledge works alongside a smart computer program.

Humans understand what is important to the company, while the AI can watch thousands of things at the same time to spot trouble.

Protecting a Global Network

In the past, hackers usually lived near the companies they attacked, but now they can strike from anywhere in the world.

Using AI lets small groups of bad actors launch big attacks without even knowing the local language.

To stay safe, security teams must work around the clock and use tools that can stop a threat the moment it appears.

Talking to the Board

Security is no longer just a job for the IT department but is now a major part of running a successful business.

Leaders are now spending more time talking to the big bosses and boards of directors about how to stay safe.

New rules and fines mean that keeping the company secure is a key part of making sure the business survives and grows.

The goal for every leader should be to bring in AI in a way that is safe and helpful for everyone in the company.

Success in the future will depend on how well leaders can combine human skill with the speed of artificial intelligence.

A major medical company called Stryker has been hit by a global cyber attack that shut down its computer systems.

The trouble started when a group of hackers from Iran targeted the company's offices in Cork and other places around the world.

Thousands of workers were unable to do their jobs because their laptops and phones suddenly stopped working.

A Digital Cleaning Tool

Experts say the hackers used a "wiper" program, which is a very mean kind of tool that deletes everything it finds.

Instead of asking for money to give the files back, this program simply erases all the data forever.

Many employees saw a special logo from the hacker group on their screens right before their devices went dark.

Trouble for Making Medicine

The attack was so big that it stopped the machines used to make important medical tools like surgical equipment.

While some factories are still trying to run, it is hard for them to keep going without their computer brains.

The company is now working with law enforcement and tech experts to try and fix the mess as fast as they can.

A Message for Other Firms

Security sources believe this attack was done to cause as much damage and chaos as possible.

It shows that even very big companies can be at risk when hackers use powerful tools for political reasons.

Other businesses are being told to watch their own systems closely to make sure they do not face the same kind of trouble.

Fixing a network after a wiper attack is like trying to rebuild a house after a fire has burned everything inside.

Telus Digital, a company that helps other businesses with customer support, has been hit by a massive data theft.

The hackers, known as ShinyHunters, claim to have taken a huge amount of information from both the company and its clients.

This attack is different because the hackers did not just break in and lock things up; they stayed hidden for months to steal data slowly.

A Smart and Patient Attack

Instead of using a virus to smash through the front door, the hackers used real logins to trick the systems into trusting them.

By moving slowly and making their traffic look like normal office work, they were able to avoid setting off any alarms.

Experts say this shows that hackers are getting much better at "blending in" rather than just breaking things.

The Danger of Stolen Identities

The group likely used fake calls and messages to trick workers into giving away their passwords.

Once they had the right keys, they could walk through the entire network and see almost everything.

Because Telus Digital works for so many other famous firms, this breach could mean that data from many different companies is now in the wrong hands.

How to Fight Back

Security leaders say companies must stop trusting every login and start checking for "strange" behavior instead of just "bad" behavior.

It is important to keep different parts of a network separate so that a hacker cannot move from one room to another easily.

Firms should also watch how much data is being moved out of the system and set alerts for when things look unusual.

The biggest lesson here is that being "trusted" by a system is now the most dangerous tool a hacker can have.