Sources

Since GPT-5.4, we’ve unified Codex and the main model into a single system, so there’s no separate coding line anymore. GPT-5.5 takes this further, with strong gains in agentic coding, computer use, and any task on a computer. — Romain Huet , confirming OpenAI won’t release a GPT-5.5-Codex model Tags: generative-ai , gpt , openai , ai , llms

(AINews) GPT 5.5 and OpenAI Codex Superapp latent.space

Spud lives!

Three reasons why DeepSeek’s new model matters technologyreview.com

On April 24, Chinese AI firm DeepSeek released a preview of V4, its long-awaited new flagship model. The model can process much longer prompts than its last generation, thanks to a new design that helps it handle large amounts of text more efficiently. Like DeepSeek’s previous models, V4 is open source, meaning it is available…

(AINews) DeepSeek V4 Pro (1.6T-A49B) and Flash (284B-A13B), Base and Instruct — runnable on Huawei Ascend chips latent.space

The prodigal Tiger returns… but is no longer the benchmarks leader.

Google will invest as much as $40 billion in Anthropic arstechnica.com

This follows a similar, but smaller, investment by Amazon just days ago.

Anthropic and NEC collaborate to build Japan’s largest AI engineering workforce anthropic.com

Anthropic is partnering with NEC to train what the two companies pitch as Japan’s largest AI engineering workforce, deepening Claude’s enterprise foothold in the country through NEC’s systems integration channels and internal developer upskilling programs.

The people do not yearn for automation simonwillison.net

Simon Willison highlights Nilay Patel’s Verge essay arguing that ChatGPT usage keeps climbing while public sentiment toward AI sours because ‘software brain’ executives reduce human experience to automatable information flows — the same instinct that has kept smart-home adoption niche for over a decade.

An update on our election safeguards anthropic.com

Anthropic published an update on the election integrity safeguards it has applied to Claude, covering policy enforcement, misuse monitoring, and partnerships meant to prevent the model from being weaponized for voter manipulation or fabricated political content.

Report: Samsung execs worried company could lose money on smartphones for the first time arstechnica.com

Samsung executives reportedly fear the smartphone division could post its first-ever annual loss, as the AI-driven memory boom drives up HBM and DRAM prices that Samsung’s own handset business has to pay, squeezing margins on Galaxy devices.

Man faces 5 years in prison for using AI to fake sighting of runaway wolf arstechnica.com

South Korean prosecutors are seeking up to five years in prison for a man who fabricated AI-generated photos of Sejong, an escaped zoo wolf that captivated the country, after sharing the fake sightings online for amusement during the manhunt.

ChatGPT’s Nano Banana bensbites.com

Ben’s Bites benchmarks ChatGPT’s new image model — informally dubbed its ‘Nano Banana’ — against popular design tools, testing how it stacks up on common creative workflows like mockups, edits, and brand-consistent generations.

AIE Europe Debrief + Agent Labs Thesis: Unsupervised Learning x Latent Space Crossover Special (2026) latent.space

Latent Space and Unsupervised Learning’s crossover episode debriefs the AI Engineer Europe conference and lays out an ‘Agent Labs’ thesis on the emerging crop of agent-focused startups; recorded before the Cursor–xAI deal landed.

Elevating Austria: Google invests in its first data center in the Alps. blog.google

Aerial image of the data center

Partnering with industry leaders to accelerate AI transformation deepmind.google

Google DeepMind partners with global consultancies to bring the power of frontier AI to organizations around the world.

8 Gemini tips for organizing your space (and life) blog.google

Black screen with illustrations including a laptop with document and email icons streaming out, a smiling person holding a phone and looking at a checklist, and a user activating Gemini Live

Quoting Maggie Appleton simonwillison.net

[…] if you ever needed another reason to learn in public by digital gardening or podcasting or streaming or whathaveyou, add on that people will assume you’re more competent than you are. This will get you invites to very cool exclusive events filled with high-achieving, interesting people, even though you have no right to be there. A+ side benefit. — Maggie Appleton , Gathering Structures ( via ) Tags: blogging , maggie-appleton

(AINews) Tasteful Tokenmaxxing latent.space

a quiet day lets us reflect on the top conversation that AI leaders are having everywhere.

References

Artificial Intelligence News artificialintelligence-news.com

GPT-5.5 is OpenAI’s most capable agentic AI model yet at twice the API price

MindStudio benchmark roundup mindstudio.ai

GPT-5.5 reached 82.7% on Terminal-Bench 2.0 … but Claude Opus 4.7 still leads SWE-bench Pro at 64.3% vs 58.6%

MindStudio long-horizon eval mindstudio.ai

GPT-5.5 required a 13.1% cleanup pass compared to just 5.8% for the latest Claude models … attributed to context drift

i10x migration guide i10x.ai

intermediate models—including gpt-5-codex and various gpt-5.1-codex variants—are scheduled for a complete shutdown on July 23, 2026

BigGo / Latent Space recap on Chronicle finance.biggo.com

memories are stored as unencrypted markdown files … Chronicle increases susceptibility to indirect prompt injection

Capwave on Cursor’s response capwave.ai

Cursor has begun training its own frontier-level model, Composer 2 … roughly 72x cheaper than GPT-5.5 Pro

Medium — ‘The 1M Context Lie’ (Aditya J) medium.com

DeepSeek-V4-Pro to operate at 1M tokens while requiring only 10% of the KV cache and 27% of the inference FLOPs used by its predecessor, V3.2

Council on Foreign Relations cfr.org

DeepSeek leads the open-weight field but still trails the leading US frontier models by roughly 3 to 6 months in raw reasoning

Tom’s Hardware tomshardware.com

DeepSeek launches 1.6-trillion-parameter V4 on Huawei chips as US escalates AI theft accusations

StartupFortune (on State Dept cable) startupfortune.com

Anthropic reported that DeepSeek and other firms utilized approximately 24,000 fraudulent accounts to generate over 16 million exchanges with the Claude model

Hacker News thread (item 47894988) news.ycombinator.com

leaderboards should include a ‘reliability score’ because frequent API errors often mask the model’s actual reasoning quality

NIST/CAISI evaluation nist.gov

DeepSeek models are significantly more vulnerable to ‘agent hijacking’ and malicious instructions compared to U.S.-based frontier models

Sources

References

Jack Sun, writing.