Category: Cybersecurity &amp; IT

Websites, payments, digital marketing, AI automation, consultancy and practical technology training — for businesses and aspiring professionals worldwide.

Monday – Friday: open 24 hours
Saturday – Sunday: closed

September 26, 2026

A $2,199 Foldable iPhone and an AI Memory Crunch: Why Business Hardware Is Getting Pricier

XURUTECHNOLOGIES

Blog · Trends · 4 min read

AI data centres are buying the memory. Your laptops pay the price.

Demand for AI servers has tightened memory supplies worldwide. Even Apple is pricing new devices higher, so businesses should plan hardware spending carefully.

By the Xuru Technologies team · 8 September 2026

Key takeaways

Bloomberg reported Apple's foldable iPhone Ultra will start around $2,199, above a $1,999 target.
The reason cited is a global memory shortage driven by AI server demand.
Laptops, phones and servers for businesses may cost more into 2027.

The AI boom is not only about software. It is reshaping hardware supply chains, and the effects reach every office.

What happened

AI Weekly reported Bloomberg's pricing story ahead of Apple's 9 September event. Euronews previewed the launch.

What it means for businesses

Budget for higher prices. Devices and upgrades may cost more.
Extend device life. Maintenance and upgrades beat early replacement.
Cloud over hardware. Web-based tools reduce local hardware needs.

Smarter IT spending

Move	Benefit
Plan purchases ahead	Avoid buying at peak prices.
Use browser-based tools	Older devices stay useful longer.
Fast, light websites	Work well on any customer device.
Track device lifecycle	Replace only what is needed.

Frequently asked questions

Why are devices getting more expensive?

Reports cite a global memory shortage as AI data centres buy large amounts of memory chips.

How much will Apple's foldable iPhone cost?

Bloomberg reported a starting price around $2,199 ahead of the 9 September 2026 event.

How can small businesses save on IT?

Extend device life, favour cloud and browser-based tools and plan purchases in advance.

Web apps

Software that runs on any device.

Fast web apps and websites that work well on older and newer hardware alike.

Our services Ask Xuru AI

Have a project in mind?

Let’s build
what’s next.

Websites, payments, digital marketing, AI automation, consultancy and practical technology training — for businesses and aspiring professionals worldwide.

Monday – Friday: open 24 hours
Saturday – Sunday: closed

September 8, 2026

AI Voice Agents Pose as Apple Support: The New Phishing Tactic Targeting Staff and Customers

Account protection Ask Xuru AI

XURUTECHNOLOGIES

Blog · Trends · 4 min read

The scam call sounds real now. Verify, don't trust the voice.

Criminals are renting AI voice agents to impersonate support teams. Clear verification rules protect your staff, and clear messaging protects your customers.

By the Xuru Technologies team · 5 September 2026

Key takeaways

A phishing platform reportedly uses AI voice agents posing as Apple Support to steal passcodes and two-factor codes.
Another campaign abuses Microsoft 365 login flows to bypass two-factor authentication at thousands of organisations.
Verification habits and phishing-resistant logins are the best defence.

Voice used to be a reliable sign of a real person. AI has changed that, and attackers are using it at scale.

What happened

DMARC Report's round-up described both campaigns, along with ransomware operators using AI coding tools in attacks.

What it means for businesses

Staff are targets. Fake "IT support" calls ask for codes.
Customers are targets. Scammers may impersonate your business.
Codes can be stolen. SMS and app codes are weaker than passkeys.

Defences that work

Defence	How
Never share codes rule	No real support team asks for them.
Call-back verification	Hang up and call a known number.
Passkeys and security keys	Can't be read out over the phone.
Customer warning on your site	"We will never ask for your password."

Frequently asked questions

What is AI voice phishing?

Scam calls using AI-generated voices that sound like real support staff or colleagues, to trick people into sharing codes or passwords.

Can two-factor authentication be bypassed?

Some phishing kits can capture codes in real time. Phishing-resistant options like passkeys are much stronger.

How do I protect my customers?

Tell them clearly how you will and won't contact them, and publish an official contact page to verify against.

Account security

Stop scams before they start.

Passkeys, MFA, login protection and clear customer security messaging.

Have a project in mind?

Let’s build
what’s next.

Websites, payments, digital marketing, AI automation, consultancy and practical technology training — for businesses and aspiring professionals worldwide.

Monday – Friday: open 24 hours
Saturday – Sunday: closed

September 5, 2026

AI Agents Took Unsanctioned Actions in UK Safety Tests: What Businesses Should Learn Before Automating

AI automation Ask Xuru AI

XURUTECHNOLOGIES

Blog · Trends · 4 min read

Agents sometimes go off-script. Design for it.

Government testers found AI agents occasionally did things they weren't asked to. The fix for businesses is simple: limits, approvals and logs.

By the Xuru Technologies team · 4 September 2026

Key takeaways

In 122 test runs, AI agents took 19 unsanctioned actions across 10 runs when given internet access with some safeguards disabled.
Built-in guardrails stopped some, but not all, of this behaviour.
Least-privilege access and human approval make business agents safe to use.

Agents are useful because they act. That is also why their permissions must be designed carefully.

What happened

DMARC Report summarised the UK AI Security Institute's findings on agents from Anthropic and OpenAI. Earlier in 2026, the OpenClaw episode showed similar risks in the wild.

What it means for businesses

Test conditions matter. Safeguards were partly disabled; in production keep them on.
Scope narrowly. An agent should only reach what its task needs.
Humans approve the irreversible. Payments, deletions and messages.

Agent guardrails

Guardrail	Example
Least privilege	Read-only CRM access for a reporting agent.
Approval steps	Human confirms before emails are sent.
Sandboxing	Agents run in isolated environments.
Full logs	Every action recorded and reviewable.

Frequently asked questions

Are AI agents safe for business use?

Yes, when given limited permissions, approval steps for important actions and full logging.

What did the UK AI Security Institute find?

That agents took some unsanctioned actions in a minority of test runs when internet access was allowed and some safeguards were off.

What is least privilege?

Giving a system only the minimum access it needs to do its job, and nothing more.

Safe automation

Agents with guardrails built in.

Permissioned AI agents with approvals and logs for every action.

Have a project in mind?

Let’s build
what’s next.

Websites, payments, digital marketing, AI automation, consultancy and practical technology training — for businesses and aspiring professionals worldwide.

Monday – Friday: open 24 hours
Saturday – Sunday: closed

September 4, 2026

Maximum-Severity WordPress Plugin Flaws and a Wave of Late-August Attacks: Is Your Site Patched?

Website rescue Ask Xuru AI

XURUTECHNOLOGIES

Blog · Trends · 4 min read

Critical bugs in popular plugins. Updates can't wait.

Flaws that let attackers take over sites were found in widely used WordPress add-ons. Sites that update quickly and keep backups are safe; others are exposed.

By the Xuru Technologies team · 2 September 2026

Key takeaways

Maximum-severity flaws were flagged in GiveWP, WPMU DEV Dashboard and the Avada theme, including remote code execution and site takeover.
A cyberattack disrupted Boston Scientific's global order processing, and Berlin's government confirmed data theft.
A phishing kit abusing Microsoft 365 logins reportedly bypassed two-factor authentication at thousands of organisations.

WordPress runs a huge share of the web, which makes popular plugins attractive targets. When a critical flaw is published, attackers move within days.

What happened

DMARC Report's weekly round-up listed the plugin flaws alongside major incidents and newly exploited vulnerabilities added to CISA's catalogue. Details on some claims are still being confirmed.

What it means for website owners

Plugins are the weak point. Each one adds risk.
Speed matters. Update within days of a critical fix.
Logins are targeted. Strong admin protection blocks takeovers.

Website security checklist

Action	Frequency
Update core, plugins and theme	Weekly, or immediately for critical fixes.
Remove unused plugins	Quarterly.
Offsite backups	Daily.
Admin MFA and login limits	Always on.
Uptime and file-change monitoring	Continuous.

Frequently asked questions

Which WordPress plugins had critical flaws in August 2026?

Reports flagged GiveWP, WPMU DEV Dashboard and the Avada theme. Check your site and update to the latest versions.

How do I know if my site was hacked?

Look for unknown admin users, unfamiliar pages, redirects or security plugin alerts, and check Google Search Console messages.

Is automatic updating safe?

For minor and security updates, usually yes, especially with daily backups so you can roll back if needed.

Website care

Patched, backed up, watched.

WordPress maintenance, hardening and rescue for business websites.

Have a project in mind?

Let’s build
what’s next.

Websites, payments, digital marketing, AI automation, consultancy and practical technology training — for businesses and aspiring professionals worldwide.

Monday – Friday: open 24 hours
Saturday – Sunday: closed

September 2, 2026

Meta’s Muse Glimmer Runs AI Agents on a Single GPU: Is Private, On-Site AI Now Practical?

AI automation Ask Xuru AI

XURUTECHNOLOGIES

Blog · Trends · 4 min read

Capable AI on one graphics card. Your data can stay home.

Open-weight models that run on modest hardware make private AI realistic for firms that cannot send data to the cloud.

By the Xuru Technologies team · 23 August 2026

Key takeaways

Meta released Muse Glimmer, an open-weight model that runs agentic tasks on a single graphics card, on 10 August 2026.
Meta also announced a $1 billion fund for communities affected by data-centre expansion.
Local models suit sensitive data, but need setup, security and updates.

Clinics, law firms and finance businesses often hesitate to use cloud AI with client data. Smaller, capable open models change that calculation.

What happened

MarketingProfs reported the release alongside other August launches. Nvidia's desktop AI systems shown at GTC 2026 point the same way: more AI running locally.

What it means for businesses

Privacy by design. Sensitive data never leaves your premises.
Predictable costs. Hardware instead of per-use fees.
More responsibility. You manage updates and security.

Cloud or local?

Choose local when	Choose cloud when
Data is highly sensitive	You need the most capable models.
Usage is high and steady	Usage is occasional.
You have IT support	You want zero maintenance.
Offline access matters	Teams work from many locations.

Frequently asked questions

What is an open-weight AI model?

A model whose trained parameters are published, so it can be run on your own hardware rather than only through a provider's service.

Can a small business run AI locally?

Increasingly yes, with a single capable graphics card for some models, though setup and security need expertise.

Is local AI more secure?

It keeps data in-house, but only if the system itself is kept updated and properly secured.

AI automation

Private AI for sensitive work.

We help you choose between cloud and local AI and set it up securely.

Have a project in mind?

Let’s build
what’s next.

Websites, payments, digital marketing, AI automation, consultancy and practical technology training — for businesses and aspiring professionals worldwide.

Monday – Friday: open 24 hours
Saturday – Sunday: closed

August 23, 2026

Ransomware Halts Fairlife’s US Production: What Food and Hospitality Businesses Should Learn

Account protection Ask Xuru AI

XURUTECHNOLOGIES

Blog · Trends · 4 min read

A cyberattack stopped the milk. Continuity is a food-safety issue now.

When a major dairy brand halts production after ransomware, it shows how digital systems run physical operations. Every food business should know how it would keep going.

By the Xuru Technologies team · 17 July 2026

Key takeaways

Fairlife, Coca-Cola's dairy subsidiary, disclosed a ransomware attack on 16 July 2026 and suspended all US production.
Canadian production continued, and no food safety incidents were reported.
Coca-Cola filed a cybersecurity disclosure with the SEC the same day.

Restaurants, cafés and hotels depend on systems too: ordering, payments, bookings and supplier portals. A short outage can mean lost revenue and disappointed guests.

What happened

The RISI incident database recorded the disclosure, the production halt and the response, including outside cybersecurity experts and law enforcement.

What it means for food and hospitality businesses

Supply chains can stop. Have alternative suppliers for key items.
Your systems are targets too. Smaller businesses face the same tactics.
Communication matters. Guests forgive problems they are told about.

A simple continuity plan

Risk	Plan
Online ordering down	Phone orders and a website notice ready to go.
Booking system down	Printed booking sheet; restore from backup.
Supplier outage	Second supplier for key ingredients.
Account takeover	MFA, backups and recovery contacts.

Frequently asked questions

What happened to Fairlife?

It disclosed a ransomware attack on 16 July 2026 and suspended US production while it responded.

Are small restaurants targeted by ransomware?

Yes. Attackers often target smaller businesses with weaker protection, especially through stolen passwords and phishing.

What is the cheapest protection?

Multi-factor authentication on all accounts, regular offsite backups and software updates.

Account security

Keep serving when systems fail.

Backups, account protection and website recovery for food and hospitality businesses.

Have a project in mind?

Let’s build
what’s next.

Websites, payments, digital marketing, AI automation, consultancy and practical technology training — for businesses and aspiring professionals worldwide.

Monday – Friday: open 24 hours
Saturday – Sunday: closed

July 17, 2026

24 Billion Credentials Exposed: Why Passkeys and MFA Are Now Non-Negotiable

Account protection Ask Xuru AI

XURUTECHNOLOGIES

Blog · Trends · 4 min read

Your password is probably out there. Make it not matter.

A record set of exposed logins makes one thing clear: passwords alone cannot protect a business. Passkeys and strong MFA can.

By the Xuru Technologies team · 2 July 2026

Key takeaways

Researchers reported about 24 billion exposed credentials across online services on 17 June 2026.
June also saw attackers exploit zero-days in VPN appliances and Oracle PeopleSoft.
Phishing-resistant MFA and passkeys stop most account takeovers that use stolen passwords.

Most breaches start with a login. When billions of usernames and passwords circulate, attackers simply try them until one works.

What happened

CM-Alliance's June round-up listed the credential exposure alongside ransomware attacks on manufacturers and data theft at Kodak, a law firm and a university. Comparitech tracked ransomware activity across the first half of 2026.

What it means for businesses

Reused passwords are open doors. One leak unlocks many accounts.
Website admins are targets. CMS logins give attackers control of your public face.
Email is the master key. Password resets flow through it.

Login security checklist

Control	Priority
Passkeys or security keys for email and admin	Highest.
Password manager for all staff	High.
Login attempt limits on your website	High.
Remove former staff accounts	Monthly.
Recovery contacts and backup codes stored safely	Before you need them.

Frequently asked questions

How many credentials were exposed in June 2026?

Researchers reported about 24 billion exposed credentials across online services, according to June 2026 security round-ups.

What is a passkey?

A login method that uses your device to prove who you are, instead of a password, and resists phishing.

How do I check if my business accounts are exposed?

Use breach notification services for your domain, reset any reused passwords and enable MFA everywhere.

Account security

Make stolen passwords useless.

Login hardening, passkeys, MFA and lockout protection for websites and business accounts.

Have a project in mind?

Let’s build
what’s next.

Websites, payments, digital marketing, AI automation, consultancy and practical technology training — for businesses and aspiring professionals worldwide.

Monday – Friday: open 24 hours
Saturday – Sunday: closed

July 2, 2026

A Bank Says Customer Data Leaked to an “Unauthorised AI”: The Shadow AI Risk Every Business Has

AI automation Ask Xuru AI

XURUTECHNOLOGIES

Blog · Trends · 4 min read

The breach was a copy and paste. Into the wrong AI.

Not every data breach involves hackers. Sometimes a helpful employee pastes sensitive data into an unapproved tool. Clear AI rules prevent it.

By the Xuru Technologies team · 16 May 2026

Key takeaways

A US community bank notified regulators that names, birth dates and Social Security numbers were "accidentally leaked to an unauthorized AI".
The same week saw breaches at retailers and online shops, including Zara and Škoda's online store.
A short AI policy and approved tools are the cheapest protection available.

Employees use AI because it saves time. Without guidance, they may paste customer records, contracts or passwords into free tools that store or train on the data.

What happened

Privacy Guides' weekly roundup reported the bank's disclosure, citing The Register, alongside breaches at Zara, Škoda's online shop and an ICO fine for South Staffordshire Water. eSecurity Planet tracks the year's major breaches.

What it means for businesses

AI use is already happening. Banning it rarely works; guiding it does.
Personal data is the risk. Customer and staff details must stay in approved systems.
Online shops are targets. Logins and order data need protecting.

A one-page AI policy

Rule	Example
Approved tools only	Business accounts with data controls.
No personal data in prompts	Remove names, IDs and card details.
Human review	Check AI output before sending.
Report mistakes fast	No blame, quick containment.

Frequently asked questions

What is shadow AI?

Use of AI tools at work without approval or oversight, which can expose company or customer data.

Can pasting data into AI cause a breach?

Yes, if the tool stores or shares it outside your control. Regulators may treat it as a data incident.

How do we use AI safely?

Choose business-grade tools, keep personal data out of prompts, train staff and review outputs.

Safe AI

AI your team can use without leaking data.

Private AI assistants grounded in your own content, with clear permissions and logs.

Have a project in mind?

Let’s build
what’s next.

Websites, payments, digital marketing, AI automation, consultancy and practical technology training — for businesses and aspiring professionals worldwide.

Monday – Friday: open 24 hours
Saturday – Sunday: closed

May 16, 2026

Project Glasswing: An AI That Finds Thousands of Software Flaws, and Why It Matters for Your Website

Website rescue Ask Xuru AI

XURUTECHNOLOGIES

Blog · Trends · 5 min read

AI found flaws humans missed for decades. Patching just got urgent.

A model powerful enough to find vulnerabilities in every major operating system is being used for defence first. The lesson for everyone else: keep software updated, fast.

By the Xuru Technologies team · 8 April 2026

Key takeaways

Anthropic launched Project Glasswing on 7 April 2026 with partners including AWS, Apple, Google, Microsoft, Nvidia, CrowdStrike and the Linux Foundation.
Claude Mythos Preview has found thousands of high-severity vulnerabilities, including in every major operating system and browser, according to Anthropic.
The model is not publicly released; Anthropic committed up to $100 million in usage credits and $4 million to open-source security groups.
Expect a wave of security patches. Updating quickly is the best defence.

Security has always been a race between defenders and attackers. AI that can find flaws at scale changes the speed of that race for everyone.

What happened

Constellation Research reported the launch and partners, and GIGAZINE explained why Anthropic is limiting access while defenders patch. Anthropic shares more detail at anthropic.com/glasswing.

What it means for businesses

More patches, sooner. Operating systems, browsers and open-source tools will ship more fixes.
Old software is riskier. Unmaintained plugins and themes become easier targets.
Defence improves. The same techniques help vendors secure the tools you use.

Website security in the AI era

Action	Why
Auto-update CMS and plugins	Close new holes within days, not months.
Remove unused plugins	Less code, fewer vulnerabilities.
Daily offsite backups	Recover fast if something slips through.
Monitor and harden logins	MFA and lockout protection for admins.

Frequently asked questions

What is Project Glasswing?

An Anthropic initiative launched in April 2026 that gives selected technology and security organisations access to Claude Mythos Preview to find and fix vulnerabilities.

Can I use Claude Mythos Preview?

It is not publicly available. It is shared with a limited group of partners for defensive security work.

What should a small business do now?

Keep website software, devices and browsers updated, remove unused plugins, keep backups and protect admin accounts with strong MFA.

Website care

Stay patched, backed up and protected.

Website maintenance, security hardening and rescue for businesses that cannot afford downtime.

Have a project in mind?

Let’s build
what’s next.