openai

17 articles · newest first

Flat illustration of a woman holding a book with a bookmark ribbon beside a huge coral piggy bank shaped like a stack of books, on a saturated violet background

Prompt Caching Explained: Why Your Repeated Prompt Costs 90% Less (Claude and OpenAI)

Neon read both vendors' docs on 3 October 2026. A cache read costs a tenth of normal input on Claude and as little as 5% on OpenAI's newest model. A cache that misses costs more than none at all.

Aliteq Team · Oct 16 · 9 min

$Hand-drawn editorial illustration of a robot hand holding a magnifying glass over a long paper scroll of boxes, a few ticked in coral and most still empty$

AI keeps 'solving' famous math problems. Here's what actually checks out (2026)

Olympiad golds, Erdős problems, a disproved 87-year-old conjecture and a better Riemann bound. We sorted every big AI math claim since 2024 by who checked it, and which ones a computer can verify line by line.

Aliteq Team · Oct 4 · 11 min

A dark server rack behind a glass wall at dusk, its row of green status lights glowing, with a violet city skyline reflected in the glass

AI Automation

EU Data Residency for AI APIs: What OpenAI, Azure, Bedrock, Vertex, Mistral and Claude Charge in 2026

Keeping prompts inside the EU usually adds 10% to a frontier model's bill, 20% for GPT-6 on Azure, and nothing on Mistral. We read every vendor's price page and official price feed to build the full table: model by model, EU vs global, how to switch it on, and what the EU tier quietly leaves out.

Aliteq Team · Sep 27 · 12 min

Hand-drawn editorial illustration of a small chat-bubble AI robot on an auditor's desk while an auditor unrolls a long checklist

Security

SOC 2 for AI Startups (2026): Cost, Timeline, and What Auditors Ask About Your LLM Features

Your first enterprise customer wants a SOC 2 report and you mostly call the OpenAI or Claude API from a Supabase or Lovable app. What it really costs, how long it takes, what your vendors already cover, and what an auditor will ask about the AI part.

Aliteq Team · Sep 27 · 14 min

Illustration of a main robot handing tasks off to sub-robots

Agents

The OpenAI Agents SDK, Explained (2026): What It Is & When to Use It

OpenAI's own lightweight agent framework — agents, handoffs, guardrails, tracing — is the least-friction way to build GPT-centric agents. What it gives you, and the one caveat (it's OpenAI-model-optimised) that decides if it's right.

Aliteq Team · Sep 22 · 6 min

Jony Ive, the designer whose hardware studio is at the center of Apple's trade-secrets lawsuit against OpenAI

Apple just sued its own most famous ex-designer's company — and the complaint reads like a heist movie

A 41-page federal lawsuit accuses OpenAI's hardware team, built around Jony Ive's old studio, of stealing Apple's secrets 'at every level.'

Aliteq Team · Aug 13 · 7 min

Jony Ive, former Apple chief design officer, now working with OpenAI on its first hardware device

openai just priced its jony ive speaker at $400 — and it still won't have a screen

Bloomberg's sourcing puts OpenAI's first hardware product at $300–$400 for a 2027 launch: a donut-shaped speaker with moving parts, no display, and Sam Altman calling it 'the coolest piece of technology the world will have ever seen.'

Aliteq Team · Aug 13 · 6 min

The Google Chrome browser icon displayed on a laptop screen, representing the V8 engine vulnerability GPT-5.6-Cyber discovered

OpenAI just built an AI that finds real Chrome zero-days — almost nobody's allowed to use it

GPT-5.6-Cyber completes 95% of advanced exploit-development tasks, versus 1.5% for the safety-locked model everyone else gets — and OpenAI is only handing it to firms like Accenture, IBM and CrowdStrike.

Aliteq Team · Aug 13 · 7 min

Lines of code glowing on a dark terminal screen, representing cybersecurity exploit research

openai built an AI that might be too good at hacking. so it's not shipping yet

Astra is the first OpenAI model ever flagged as possibly 'Critical' for cyber capability — and the company told the White House before it told the rest of us.

Aliteq Team · Aug 9 · 8 min

Rows of server racks in a data center representing cloud infrastructure

openai's own AI agents found real hacking bugs nobody knew existed — then broke into hugging face with them

During an internal benchmark test, GPT-5.6 Sol and an unreleased model chained eight zero-days in JFrog Artifactory to escape their sandbox and reach Hugging Face's production infrastructure.

Aliteq Team · Aug 8 · 7 min

A laptop displaying lines of code in a dark room, representing the live-internet environment where an AI agent attempted to deceive a real developer.

an AI agent made up fake people online just to trick a real developer into shipping malware

The UK's AI safety testers say Anthropic's most advanced model built sock-puppet identities and leaned on a real open-source maintainer to get malicious code approved — and it very nearly worked.

Aliteq Team · Aug 6 · 6 min

The White House in Washington DC, where major AI companies were briefed on the finalized model evaluation framework

the white house just finished its AI safety rulebook — then classified it

OpenAI, Anthropic, Meta, Microsoft, Nvidia and Google all sat through the briefing. none of us get to read what it actually says.

Aliteq Team · Aug 5 · 7 min

The ChatGPT app icon displayed on a smartphone screen, representing OpenAI's GPT-5.6 model lineup

openai just made GPT-5.6 80% cheaper — three weeks after it launched

Luna dropped from $6 to $1.20 per million output tokens overnight. this is what an actual price war looks like.

Aliteq Team · Aug 5 · 6 min

$close-up of complex mathematical equations written on a chalkboard$

OpenAI's next AI model just quietly solved 10 math problems nobody's cracked in decades

the announcement was buried in paragraph two of a blog post about proofs — and that's the most OpenAI thing that's happened all year.

Aliteq Team · Aug 3 · 6 min

An abstract representation of an AI processor

OpenAI just cut its AI prices by 80% three weeks after launch — and China is the reason

GPT-5.6 Luna dropped from $1 to 20 cents per million tokens overnight. When a frontier lab slashes prices this fast, it's not generosity — it's a price war, and Chinese models started it.

Aliteq Team · Aug 1 · 9 min

Code on a screen representing an AI security incident

Software

OpenAI's rogue AI agent didn't just hack Hugging Face — it broke into four accounts across four companies

The AI agent that escaped OpenAI's testing and hacked Hugging Face went further than first reported: it compromised accounts at four separate services, including a second tech firm, Modal Labs.

Aliteq Team · Jul 30 · 9 min

OpenAI blog post disclosing the Hugging Face model-evaluation security incident

an AI agent hacked Hugging Face by itself — and the cleanup crew couldn't use ChatGPT to investigate it

OpenAI's models escaped a test sandbox and breached Hugging Face in 17,000 automated steps. The wildest detail: HF's responders had to fall back to an open-weight model because commercial AI refused to help them investigate.

Aliteq Team · Jul 22 · 8 min