CEO of a nationally award winning IT consulting firm, appear as a regular on Fox 29 news, love technology, new ideas, and am inspired by entrepreneurs.
Moorestown, NJ
Joined August 2010
- Tweets4.1K
- Following1.6K
- Followers6.8K
- Likes5K
A critical GitLab flaw, rated 10.0 out of 10, could let attackers run code on self-managed servers without logging in, and it reaches back 12 years of releases. If your team hosts its own GitLab, check the advisory and patch tonight. #Cybersecurity
Phishing is getting personal. One attacker posed as 30+ different buyers, sending real estate agents unique property inquiries and even video meeting invites. If a new client pushes a link before you've ever spoken, verify them another way first. #Cybersecurity
Publishing a yearly public benefit report, with clear pillars and an honest look at progress on each, is how AI companies earn trust. Transparency like this helps every business feel better about putting these tools to work.
Anthropic is a Public Benefit Corporation. Today, we’re publishing our annual Public Benefit Report. This report shares our five pillars and takes stock of the actions we’re taking to advance them.
Read the full report at this page: anthropic.com/public-benefit…
AI agents are only as safe as the limits around them. Setting clear boundaries on what an agent can touch, and being able to stop it when it steps outside them, is what will let businesses move agents from pilots into real work.
PHP admins: the patch for the two leaked PHP flaws is due this morning, and exploit code is already public. The one-line config workaround shared overnight isn't enough on its own, so plan to upgrade PHP today on WordPress and every other PHP site you run. #Cybersecurity
OpenAI now offers 12-hour prompt caching in its API. AI agents that reuse the same instructions and tools all day can pay the lower cached-input rate, even when requests are hours apart. If your team runs assistants on a schedule, that's an easy cost win. #AI
Supply chain check: attackers slipped backdoored releases of a cloud provider's CLI onto GitHub to steal developer secrets. Pin your tool versions, verify checksums or signatures before installing, and rotate keys on any machine that pulled a bad build. #Cybersecurity
Scorpion stings hit about 1.2 million people a year, and antivenom is often costly or hard to get. Seeing Claude help design protein binders against those neurotoxins is the kind of AI progress that makes me optimistic. Early step, big potential.
New on our Science Blog: designing protein binders against scorpion neurotoxins.
Scorpions sting about 1.2m people every year, and antivenoms are often costly or in short supply. Claude has made an early step toward finding new treatments.
anthropic.com/research/claud…
A lot of critical software still runs on COBOL, Fortran, and Visual Basic, and the people who know those systems are getting harder to find. AI that can help build and migrate that code, paired with solid testing, gives banks and businesses a real path to modernize.
Our latest model advances bring stronger capabilities for building and migrating software in low-resource programming languages such as COBOL, Fortran, and Visual Basic.
These languages power software that banks, governments, and businesses depend on. Progress here means understanding decades of business logic and helping developers modernize those systems while preserving what works.
We’re building these capabilities around the challenges developers and enterprise customers bring us. If your team is working on these systems, tell us which languages and workflows we should tackle next.
openai.com/index/low-resourc…
Developers: Go just shipped an urgent security update that blocks 8 certificate authority keys, and the filter can't be turned off. Apps rebuilt on the new version may reject some certificate chains, so test outbound connections after upgrading. #DevOps
One attacker reportedly used open-source AI agents to break into 100+ e-commerce and payment sites at about $50 per breach. Cheap, automated attacks make every site a target. Keep plugins patched, require MFA on admin logins, and watch for new admin accounts. #Cybersecurity
Phishing now targets your AI inbox assistant too. Barracuda found emails hiding invisible instructions that tell AI helpers to change vendor bank details. If an assistant can act on your mail, limit what it can do and require a human OK on any payment change. #Cybersecurity
Treating AI agents like insiders, with security controls that sit outside the model, is the right instinct. Continuous risk checks and permissions that adapt on the fly are how companies will deploy agents with confidence.
I recently wrote about why we need to treat models as insider risks, with security controls sitting outside the model itself.
Today, we're putting that principle into practice by expanding Project Perception with new capabilities to help continuously assess agent risk, adapt permissions, and contain threats.
blogs.microsoft.com/blog/202…
An AI model found an attack that cut HAWK's effective security in half in about 60 hours, and it was pulled from NIST consideration. That's the case for crypto agility: know where your encryption lives and make sure you can swap algorithms without rebuilding everything.
AI is testing the limits of post-quantum cryptography.
An Anthropic AI model found an attack that halved HAWK's effective security in just 60 hours, prompting its withdrawal from NIST consideration.
Itan Barmes of QIZ Security explains why cryptographic agility now matters more than ever.
Read: thehackernews.com/expert-ins…
Heads-up for IT teams: Microsoft says some October Windows Server updates are breaking Exchange sign-in and mail flow. Before pushing this month's patches to mail servers, test on one box first, keep a rollback plan ready, and watch for Microsoft's fix. #ITOps
Reports today say attackers are taking over exposed Plex Media Servers, and exploit code is already public. If you run Plex at home or at the office, update it now and keep it off the open internet unless you truly need it. #Cybersecurity
Even trusted software can carry a threat. A popular paid WordPress plugin shipped a backdoored update this week. If you run WordPress, know who manages your plugins, keep backups, and rotate passwords and API keys when a vendor says to. #Cybersecurity
Attackers tampered with AOMEI Backupper's update servers in 5 to 10 minute bursts and delivered a backdoor to PCs that checked for updates, across 11 countries. AOMEI says its servers are fixed, but infected machines stay compromised. If your team runs AOMEI, review which machines updated recently and rebuild any that pulled a bad update. #Cybersecurity
A BeyondTrust Remote Support zero-day (CVE-2026-102764) gave attackers root and a path into internal systems. One host fired alerts instantly, then an automated triage tool closed them as false positives. Patch Remote Support now, and keep a human in the loop on high-severity alerts before automation dismisses them. #Cybersecurity
Plain-language instructions in, quantum circuits out that beat the best published designs, sometimes by an order of magnitude. Kudos to Anthropic's STEM Fellows for showing how Claude Mythos 5 can widen what a small research team can explore.
New from our STEM Fellows Program: Designing more efficient quantum circuits with Claude.
Given plain-language instructions, Claude Mythos 5 designed quantum circuits that improve on the best published ones, sometimes by an order of magnitude or more.
More efficient circuits reduce the resources needed by quantum algorithms, which could bring quantum computing applications in chemistry and materials science within reach sooner.
Read more: anthropic.com/research/desig…
