AI is moving deeper into science, cybersecurity, and healthcare. OpenAI's GPT-6 Astra is improving how machines interpret laboratory experiments. Meanwhile, researchers discovered over 800 AI hacking systems exposed online without passwords. And in healthcare, a pulmonologist is using AI to help detect lung cancer. These developments reveal three sides of the same technological shift: smarter research, growing security risks, and new medical possibilities. The bigger question is whether our ability to manage AI can keep pace with its capabilities.

In today’s post:

• Stories

• OpenAI Tests Whether AI Can Think Like a Scientist

• Hundreds of AI Hacking Systems Left Open Without Passwords

SPONSORED BY

It's time to set up your holiday streaming campaign

Are you ready to reach holiday shoppers on CTV? For many streamers, the season is already underway.

Last year, according to Roku internal data, holiday content searches on Roku saw an 86% increase during October. And in November, those searches rise an average of 39% weekly. This surging holiday spirit is mirrored by consumer activity, as about half of Roku streamers shopped Black Friday and Cyber Monday deals in 2025.

Meeting those shoppers is easier than you might think. Begin with the seasonal outcome you hope to drive: awareness, online sales, leads, store visits, or another measurable goal. From there, define your audience, upload your creative, set the budget and campaign dates, and establish measurement before launching. You can even create a shoppable ad unit that lets streamers buy with a click of their remote.

Our holiday playbook offers a month-by-month guide to reaching shoppers during this all-important time of year.

First-time spenders get a $5,000 match on the first $5,000 spent with code BEEHIIV5K. Terms apply.

Launch your campaign today.

What’s Trending Today

AI MEDICARE

AI is entering the fight against lung cancer

Doctors are beginning to use AI to examine lung scans. WKRN News 2 reported a new example on November 8, 2026. A pulmonologist is using AI to help detect lung cancer. The technology assists with reading medical scans. It's a small glimpse into a potentially important shift. AI may change how doctors investigate serious diseases.

• AI is being used to examine lung scans. The report highlights a pulmonologist incorporating AI into cancer detection.

• Medical imaging is becoming an important application for AI. Algorithms can help analyze scan patterns that deserve closer examination.

• Early detection could make a meaningful difference. Finding lung cancer sooner can create more treatment opportunities.

• AI could provide doctors with additional information. Its role is to support clinical interpretation, not replace medical judgment.

• Accuracy remains essential. A suspicious scan finding doesn't automatically establish a cancer diagnosis.

• Human oversight still matters. Doctors must evaluate findings alongside patient histories and other medical evidence.

• Real-world effectiveness needs careful measurement. The WKRN report doesn't provide accuracy rates or patient outcomes.

The interesting question isn't whether AI can read scans. It's whether AI can help doctors make better decisions. Healthcare doesn't need AI to replace experienced physicians. It needs tools that make their expertise more effective. Cancer detection seems like a promising place to start. But promising technology must prove itself through patient outcomes. A faster scan interpretation means little without reliable results.

RESEARCH

AI just got better at reading lab experiments

GPT-6 Astra is learning something important: when not to trust the data. OpenAI introduced BioBasicsBench on November 6, 2026. The benchmark tests AI agents on 100 everyday laboratory tasks. These include interpreting images, analyzing experiments, and spotting mistakes. The results reveal significant progress. But they also expose why scientific judgment remains difficult for AI.

• GPT-6 Astra achieved 66.7% on BioBasicsBench at the highest shared reasoning setting. GPT-5.6 Sol scored just 43.8%. That's a 22.9 percentage-point improvement.

• The benchmark uses real experimental evidence. Scientists supplied microscopy images, DNA gels, sequencing files, and laboratory recordings. These aren't polished examples selected for research papers.

• Astra became better at recognizing uncertainty. In one experiment, overlapping fluorescence signals made cell viability unclear. Astra identified the ambiguity instead of drawing an unsupported conclusion.

• Small details made a big difference. Astra spotted an incorrectly labeled antibiotic plate. An earlier model missed it and trusted an invalid experimental control.

• Context mattered as much as visual accuracy. Astra correctly interpreted changing microscope images as different focal depths. An earlier model searched for a biological explanation.

• Better reasoning didn't eliminate fundamental mistakes. Astra selected incorrect colonies despite identifying distinct DNA band patterns. It also overlooked contradictory observations in microscopy images.

• Even basic counting remains unreliable. Astra overcounted colonies by approximately 12% in one experiment. Such errors can distort calculations and influence subsequent research decisions.

The bigger story isn't just improved benchmark performance. It's the growing importance of knowing when evidence is insufficient. Scientific AI needs more than accurate answers. It needs the discipline to question its own conclusions. Astra's progress is meaningful, but its mistakes reveal something important. Recognizing evidence isn't the same as interpreting it correctly. The next breakthrough may depend on that distinction.

CYBER SECURITY

Hundreds of AI hacking tools were left open to the internet

The same AI tools used to break into systems became security risks themselves. Security researchers at Censys uncovered a surprising security problem. More than 800 deployments of Artex were publicly accessible without passwords. Artex is an AI-powered tool designed to find security weaknesses. But its exposed installations could also help outsiders launch cyberattacks. The discovery reveals an uncomfortable irony in AI security.

• Over 800 Artex deployments were exposed online. Censys discovered these systems required no login credentials. Anyone who found them could potentially access their capabilities.

• Artex can automate several stages of cyberattacks. Microsoft previously described its agents as handling reconnaissance through data theft. These operations require relatively little human involvement.

• The tool has already appeared in major cyberattacks. CrowdStrike linked Artex to attacks targeting South Korea's financial sector. Researchers also found evidence involving payment and gaming companies.

• An October attack targeted ConnectWise customers. Microsoft reported hackers using Artex to access cloud environments. The attacks moved quickly, leaving defenders little response time.

• Basic security failures enabled many attacks. Weak passwords and missing multi-factor authentication created opportunities for hackers. Sophisticated AI wasn't always necessary to exploit these weaknesses.

• Exposed tools also revealed information about their operators. One attacker accidentally left an accessible directory containing attack infrastructure. Researchers could trace activity extending beyond the initial targets.

• Artex plans to restrict future software releases. The project announced a move away from open source. However, existing code copies remain available on public repositories.

AI is making cybersecurity more powerful. But it's also making familiar security mistakes more consequential. The most interesting part isn't that AI can automate hacking. It's that powerful systems still depend on basic security practices. A missing password can undermine even sophisticated technology. And automation makes the consequences harder to contain. Perhaps cybersecurity's next challenge isn't building smarter defenses. It's ensuring we don't forget the fundamentals.

Free Guides

My Free Guides to Download:

🚀 Founders & AI Builders, Listen up!

If you’ve built an AI tool, here’s an opportunity to gain serious visibility.

Nextool AI is a leading tools aggregator that offers:

• 500k+ page views and a rapidly growing audience.

• Exposure to developers, entrepreneurs, and tech enthusiasts actively searching for innovative tools.

• A spot in a curated list of cutting-edge AI tools, trusted by the community.

• Increased traffic, users, and brand recognition for your tool.

Take the next step to grow your tool’s reach and impact.

Reach 150,000+ READERS:

Expand your reach and boost your brand’s visibility!

Partner with Nextool AI to showcase your product or service to 140,000+ engaged subscribers, including entrepreneurs, tech enthusiasts, developers, and industry leaders.

Ready to make an impact? Visit our sponsorship website to explore sponsorship opportunities and learn more!