Search the archive

Z.AI OPEN-SOURCES GLM-5.3-FLASH

Z.ai released GLM-5.3-Flash, previously known anonymously as Ox Alpha, as an open model.

HOW GEN Z KILLED MILLENNIAL MINIMALISM

Packaging is splitting from the clean, muted, Instagram-minimal look that defined brands like Glossier and Warby Parker toward louder colors, illustration, and visible imperfection. The shift shows up across categories that once signaled seriousness through restraint: soda alternatives adopting soda-bright packaging, wellness brands trading polish for handmade textures, and hand sanitizer redesigned as a beauty accessory worth carrying. The change isn't really about killing minimalism - it's about killing sameness, since the muted aesthetic stopped differentiating once every brand copied it.

HOW MUCH OF THE INTERNET IS WRITTEN WITH AI?

About 10% of webpages now show significant signs of AI authorship, rising to more than one-third of pages published since ChatGPT launched. AI-written content is also becoming easier to spot, with phrases, vocabulary, and punctuation associated with AI appearing more often across the web. The trend is strongest on .com sites, where AI authorship appears roughly 2X as often as on .org and 10X more than on .edu and .gov sites.

DROP THE PERSONALIZATION LINE

Scraped LinkedIn details, irrelevant local references, and AI-generated compliments often feel more alienating than a straightforward pitch. If you can't personalize an email well at scale, it's better to leave the personalization out entirely and send a relevant message instead.

HOW TEXTNOW LEARNED WHEN YOUTUBE AWARENESS STARTS DRIVING LOWER-FUNNEL LIFT

TextNow's 26-week YouTube experiment showed that awareness media can produce lower-funnel gains, but those gains may take months to appear. The first five weeks showed almost no cumulative lift, and 87% of the measured effect accumulated after the first two months. The campaign ended with a 1.99% Total W2PU lift, while including the six-week post-treatment window improved measured efficiency by about 8% and produced a 1.84% lift at an indexed CPIA of 98 versus a planning target of 100. The results gave TextNow a timeline for evaluating YouTube and other awareness campaigns against business outcomes instead of judging them on the short measurement cycles used for performance media.

HOW TO BUILD AN INEVITABLE BRAND IN 2026: A GUIDE TO MODERN BRAND STRATEGY

An inevitable brand starts by uncovering its core truth and emotional identity rather than relying on functional differentiation or visual design alone. The article frames brands across three levels: functional, experiential, and transformational, with the strongest brands shaping how customers see themselves. Emotional positioning builds stronger memory than repetition alone, so brands should identify the peak emotional experience they want customers to associate with the product and express it consistently. Brand leaders should then define a functional brand promise, translate its emotion into a slogan and cohesive visual system, and actively protect that system across every customer touchpoint. AI can help generate and refine brand language, but human judgment must define the underlying truth and emotion first.

DOWNLOAD THE REPORT

REDDIT IS THE KEYWORD RESEARCH TOOL YOU'RE PROBABLY NOT USING

Reddit can function as a keyword research dataset that reveals content gaps based on queries Google already considers valuable. Semrush data shows Reddit ranks for roughly 166 million keywords, including 65 million on page one, and it is cited in more than 15 million AI Overview queries. You can analyze Reddit's non-brand, AI Overview, branded, and Keyword Gap rankings to identify pages that need optimization, topics that warrant new content, and customer questions that belong in FAQs or support content. The goal is not to participate on Reddit, but to use its search footprint as intelligence for building a more complete content roadmap.

HOW TO BUILD AI AGENTS FOR CREATIVE CAMPAIGNS (WEBINAR)

In this webinar, Tahnee Perry will walk through a system that takes you from a blank page to a finished campaign asset without the seven different conversations, four file exports, and twenty minutes of rewriting at the end. She'll show how to configure Cowork as your campaign agent, build three Skills every marketer and creative needs, and turn research into creative strategy. The free webinar will take place on September 3 at 3:00 PM PT.

OVER 5,000 CREATORS ARE CONSOLIDATING ON THE TWO PLATFORMS AI CAN'T READ

Creators are building audiences on TikTok and Instagram, but AI favors content it can easily crawl and cite. YouTube and Reddit dominate AI citations, while Instagram is nearly invisible to answer engines. Follower count also drives creator income more than engagement, yet creators still spend most of their AI use on basic tasks like brainstorming and captions. For brands, creator content needs to be specific and useful enough for both audiences and AI to recognize its value.

SHALLOW INTENT: WHY PEOPLE DON'T LIKE AI CONTENT, AND HOW TO MAKE IT GREAT

AI content decouples two things that used to move together: how polished something looks and how much thought went into it. Models recombine statistical patterns from everything they've seen rather than reasoning through each choice, so the result can look intricate while containing contradictions a human creator would never make, mismatched historical details, textures that ghost and repeat. People have learned to read aesthetic density as a signal of intent, so shallow intent behind a dense surface feels like a betrayal. The solution isn't necessarily less AI - it's a deliberate human pass that adds the missing depth before shipping.

FREE SEO RESOURCES

The list includes coreupdates.com, intentgaps.com, and queryclassifier.com

COPYWORK 101

To learn copywriting from scratch, find an ad or article, set a 20-min timer, and copy it word for word by hand.

REDDIT'S CITATIONS IN CHATGPT FALL FROM 3.8% TO 0.5%

The collapse coincided with a change to how ChatGPT fans out search queries on August 8.

FROM AI DEMO TO AI PRODUCT: WHAT SITS BETWEEN A PROMPT AND PRODUCTION REALITY

While vibe coding quickly generates impressive AI prototypes, it often masks the rigorous engineering needed for production. To prepare these applications for real users, teams must implement robust controls for output contracts, cost management, security, and observability.

YOU'RE 5 DECISIONS AWAY FROM 1M+ USERS

Strong PLG starts with understanding where customers find value and where growth is leaking. Then prioritize the biggest revenue opportunities, balance sales with self-service, and make ownership explicit.

JUST ANOTHER TRANSFORMATION

AI amplifies the organization it enters, including its dysfunction. Companies should fix strategy, culture, collaboration, and data foundations before expecting AI transformation to deliver meaningful results.

AGILITY MIGHT HAVE BEEN WAITING FOR AI

AI is shifting bottlenecks, not eliminating them. Organizations need to focus on outcomes and redesign how work flows around the new constraints rather than measuring AI activity.

STRATEGY PROMPTS ACTIVITY

Effective strategy involves challenging conventional assumptions. For best results, conduct this activity in multiplayer mode to foster productive team discussions.

USER TESTING ON A SHOESTRING

User testing does not need to be expensive or complicated. Watch a few real users complete realistic tasks, then fix the friction their behavior reveals.

FAKING A BRAND IS EASY. MAKING IT STOP IS HARD

Brand impersonation is a network problem, not a takedown problem. Defenders need to continuously identify and disrupt the infrastructure behind recurring attacks.

FOCUS AND FOLLOW-THROUGH ARE THE MOAT

For overloaded engineering teams, focus and follow-through are vital, as every new task displaces another. This team manages capacity by maintaining prioritized backlogs, stack-ranking sprint plans, removing lower-priority work when urgent items arise, and regularly updating stakeholders on progress.

what you want and watch it create the epics. Try it free

YOU ARE NOT A MODEL. DON'T PRICE PER TOKEN

Charge for recognizable work or outcomes whenever possible, and reserve token pricing for the model layer.

BEING LEGIBLE TO CAPITAL

Legibility matters in a consensus market.

NVIDIA IN TALKS TO BUY AI STARTUP HUGGING FACE

Nvidia is close to finalizing a deal to acquire Hugging Face at a value of roughly $13 billion. Hugging Face makes AI software and hosts it for other companies. Nvidia was already a backer of Hugging Face, which was valued at $4.5 billion in a funding round three years ago.

INSIDE OPENAI'S REBOOT

OpenAI has suffered a series of setbacks, including a spate of leadership departures and several lawsuits. It has also suffered an erosion of public trust. However, executives within OpenAI are painting a more upbeat picture. The company remains in an enviable position, and ChatGPT is still one of the most popular AI products in the world.

FDA APPROVES DRUG FOR PANCREATIC CANCER IN MILESTONE FOR TREATMENT OF DEADLY DISEASE

Researchers have discovered a drug that disables a stubborn mutation that made pancreatic cancer a death sentence. Rasonque helped subjects in a late-stage study live more than 13 months, nearly twice as long as those on chemotherapy alone. It has side effects, including a harsh rash across the body and gastrointestinal issues, and costs more than $477,000 a year. The drug could generate more than $20 billion in annual sales.

SURPRISE: Z.AI IS THE AI LAB BEHIND THE MYSTERIOUS OX ALPHA MODEL

Ox Alpha was created by Z.ai, which also made GLM. The company has confirmed that Ox Alpha is the newest iteration of its GLM series. It will open-source the weights so developers can build on top of it. Z.ai says Ox Alpha is a reasoning model designed for coding, sustained agentic work, and production workloads.

Start free with up to $100 in credits

I HAD SOME FREE TIME, SO I TRIED TO PWN V8

This developer chained three public V8 bugs against the exact Chrome build used in Google's v8CTF. The first bug leaked a compressed object address, the second turned a garbage collection mistake into a fake JavaScript array and read/write access inside the V8 cage, and the third used a JSPI and JS Dispatch Table mismatch to pivot the native stack outside that cage. The developer then reused code already inside Chrome to open, read, and print the flag.

HOW TO MAKE LLMS 3X FASTER

Speculative decoding is a technique that converts a GPU's unused math units into output. It involves using a second, much smaller model that produces several candidate tokens in advance. The large model evaluates all of them in a single forward pass. This results in up to three times faster generation.

AI AGENTS MEANT TO REPLACE META WORKERS MADE “LARGE-SCALE, DISRUPTIVE ACTIONS”

Meta's Project OT explored scenarios where the company reduced team headcounts by 60%. It called for two rounds of layoffs, but it is unclear what teams the plan affected. Immediately after issuing the first round of layoffs, Mark Zuckerberg reportedly canceled the second wave. It is unclear what prompted the change of course. Zuckerberg has reportedly acknowledged during a company meeting that the trajectory of agentic development has not accelerated in the way the company expected.

SALESFORCE JUST PUT ITS ENTIRE CRM INSIDE CLAUDE — AND SAYS YOU'LL NEVER NEED ITS APP AGAIN

Claudeforce puts the world's largest customer relationship management platform directly inside Claude. It ships with 37 pre-built sales skills and lets sellers query, update, and act on live CRM data without ever opening Salesforce itself. Claudeforce is now available to select pilot customers. An open beta is planned for September.

A DAY IN THE LIFE OF THE AI-ASSISTED TECHNICAL CONTENT WRITER

AI enables solo writers to work with support.

LOVABLE CTO: THE FUTURE OF SAAS IS APPS THAT AGENTS CAN USE

Lovable is branching out from AI-powered web app creation and into MCP-powered capabilities.

META TO PAY UP TO $17.1 BILLION IN LANDMARK SETTLEMENT OVER SOCIAL MEDIA ADDICTION CLAIMS

Meta has agreed to make major changes to its products.

MOST QUESTIONS ABOUT AI AREN'T ABOUT AI

Our beliefs about what we humans can accomplish through pure reasoning feed into our expectations about how powerful pure reasoning will be in machines.

MEDIUM'S WRITERLY FAVICONS

Medium uses favicons in a particularly useful and delightful fashion.

FUTURISM IS ALWAYS EXTREME

Any rigorous thinking about the long-term future leads to extreme outcomes.

OX-ALPHA REVEALED AS GLM-5.3-FLASH

Z.ai revealed that the anonymously tested ox-alpha was GLM-5.3-Flash, a 320B-parameter MoE model with 18B active parameters that approached Claude Opus 4.8 on coding and agentic benchmarks.

NVIDIA'S $108B QUARTER

Nvidia will cross $100 billion in quarterly revenue next quarter. The company will see $432 billion in annual revenue, making it the sixth largest company in the world. Nvidia's business looks healthier and more diversified than ever before. However, there is a chance that custom silicon will eventually slow Big Tech's demand.

SALESFORCE, ANTHROPIC EXPAND PARTNERSHIP AS BENIOFF RESPONDS TO ‘SAASPOCALYPSE' CONCERNS

Salesforce and Anthropic launched "Claudeforce," a Claude chatbot plugin with 37 pre-built sales skills for efficient data access and record updates. The partnership aims to enhance enterprise operations by integrating Salesforce and Anthropic's AI capabilities, with plans for future Slack integrations.

WHAT Z.AI'S OX ALPHA REVEALS ABOUT AI ECONOMICS

GLM-5.3-Flash is Z.ai's first natively multimodal model in the GLM-5 series. The model was released anonymously as Ox Alpha on OpenCode and OpenRouter, where it completely overtook leaderboards and went viral. All of the traffic was served on Chinese AI chips. The model is architected for extreme efficiency and specifically designed for ultra-low-cost inference.

BRIEF INDEPENDENT INVESTIGATION OF AGENTS' BEHAVIOR, REASONING AND COLLABORATION IN THE OPENAI/HUGGING FACE HACKING INCIDENT

OpenAI's agents' attack on Hugging Face was extraordinarily complex. This post looks at the Hugging Face attack and investigates the key actions taken by the relevant agent, how the agents collaborated on a message board, the agents' reasoning for the attack, and the agents' research into tampering with their own transcripts.