Skip to content

#其他

0 today

Sep 23Wednesday

Hacker News front page

Pentagon says overreliance on AI contributed to missile strike on Iran school

A Pentagon probe found that overreliance on AI targeting contributed to a 2025 US missile strike that hit an Iranian school. The system mislabeled the school as a military site, and human operators did not override the machine's call within an 11-second decision window. The report names Palantir's Maven system and Google AI tools, though the post doesn't spell out exactly which component failed. This wasn't autonomous firing—it was human-machine teaming where humans deferred to the machine.

Why it matters: Pentagon investigators reconstructed the strike on a school in Minab, Iran, and traced the deaths of 123 children to bad intelligence, outdated satellite imagery and overreliance on AI.

AI HOT (Curated Pool)

Anthropic Releases Claude Opus 5.5: Fable 5.1-Level Performance at 40% Lower Running Cost Than Opus 5

Anthropic launched Claude Opus 5.5, the first model in its Claude 5.5 family. The team says it matches Fable 5.1 on most work while costing 40% less to run than Opus 5. It leads Anthropic's internal benchmarks on agentic coding, computer use, and knowledge work. It's not a clean sweep—GPT-6 Astra still leads on Terminal-Bench-Science and AutomationBench. Pricing is $4 per 1M input tokens, $20 per 1M output tokens, and cache reads drop to $0.20, a 60% cut that matters most for agentic and coding costs. Output is over 30% faster than Opus 5, with a fast mode offering 2.5x speed. The model is API-only, no open weights. One early tester migrated 680,000 lines of code in under a day.

Why it matters: Anthropic's Opus 5.5 arrives with claims of Fable 5.1-level work on most tasks and 40% lower running cost than Opus 5 on typical loads at default settings.

AI HOT (Curated Pool)

GPT-6 Sol and GPT-6 Luna land on Arena, API pricing 50% below GPT-5.6 promo rates

OpenAI dropped GPT-6 Sol and GPT-6 Luna on Arena, both built on GPT-6 Astra tech. The pitch is faster, cheaper inference with better caching for high-volume workloads. API pricing is 50% below GPT-5.6's promotional rate. The post doesn't disclose benchmark scores or latency numbers, so I'd wait for third-party benchmarks before getting excited.

Why it matters: OpenAI put GPT-6 Sol and Luna on Arena at API prices 50% below GPT-5.6's promotional rate, with caching and inference tuned for high-throughput work.

AI HOT (Curated Pool)

Sam Altman says GPT-6 Sol and Luna have no competition on per-task pricing

Sam Altman posted that GPT-6 Sol and Luna have no competition when measured by per-task pricing. He claims big jumps over the 5.6 series in intelligence, alignment, work output, coding, and computer use, with per-token price halved and even lower per-task cost. The post doesn't disclose specific benchmarks or pricing figures—I'd wait for third-party testing before taking it at face value.

Why it matters: Altman claims GPT-6 Sol and Luna have no rivals on task-based pricing, with no benchmarks or prices to back it up.

AI HOT (Curated Pool)

OpenAI launches GPT-6 Sol and GPT-6 Luna, API pricing 50% below GPT-5.6 promo rates

OpenAI released GPT-6 Sol and GPT-6 Luna, both built on GPT-6 Astra tech and aimed at cheaper, faster high-volume workloads. API pricing is 50% lower than GPT-5.6 promotional pricing, driven by more efficient caching and inference. Sam Altman reposted the announcement and called the character designs cute. The post doesn't disclose benchmark scores, latency figures, or regional availability.

Why it matters: OpenAI released GPT-6 Sol and GPT-6 Luna at API prices 50% below GPT-5.6's promotional rate, pitched at cheap, fast work on high-volume tasks.

AI HOT (Curated Pool)

OpenAI GPT-6 Sol and Luna land on OpenRouter at half the price

OpenRouter just listed two new OpenAI models: GPT-6 Sol and GPT-6 Luna. Pricing is half that of the previous GPT-5.6—Sol at $2/M input and $10/M output, Luna at $0.10/M input and $0.50/M output. On AutomationBench, both beat the prior best score while costing less per task. The post doesn't disclose exact scores or latency figures.

Why it matters: OpenAI listed GPT-6 Sol and Luna on OpenRouter at half the price of the previous GPT-5.6, and both beat the prior generation's best AutomationBench score at lower cost per task.

AI HOT (Curated Pool)

GPT-6 Sol and Luna: half the price, same Intelligence Index

Artificial Analysis tested OpenAI's GPT-6 Sol and Luna. Both cost half as much as their GPT-5.6 equivalents: Sol at $2/$10 per million input/output tokens, Luna at $0.10/$0.50. The Intelligence Index matches GPT-5.6, so you're getting the same capability for less money. The post doesn't disclose evaluation dimensions or latency figures.

Why it matters: Artificial Analysis found GPT-6 Sol and Luna priced at roughly half their GPT-5.6 counterparts with matching Intelligence Index scores, giving a direct price reference for model selection and cost planning.

AI HOT (Curated Pool)

OpenAI launches GPT-6 Sol and Luna, API pricing 50% lower than GPT-5.6

OpenAI's developer account announced GPT-6 Sol and Luna, with API pricing cut to half of GPT-5.6. Sherwin Wu added specifics for Luna: $0.10 per 1M input tokens and $0.50 per 1M output tokens, noting that pricing will soon need to switch to per-billion-token units. The post doesn't disclose Sol's per-token price or how the two models differ in capabilities.

Why it matters: OpenAI's flagship swap comes with a 50% API price cut, and Luna lands at $0.10 per million input tokens.

AI HOT (Curated Pool)

OpenAI rolls out GPT-6 Sol and GPT-6 Luna to ChatGPT Work and Codex

OpenAI announced the rollout of GPT-6 Sol and GPT-6 Luna for Plus, Pro, Business, Enterprise, and Edu users, available now in ChatGPT Work and Codex. The post doesn't disclose model specs, pricing changes, or performance vs. GPT-5—hold for benchmarks.

Why it matters: Shows which paid tiers and which entry points, ChatGPT Work and Codex, can reach the two new models today.

TechCrunch · AI

OpenAI launches GPT-6 Sol and Luna, boasting lower cost and fewer mistakes

OpenAI followed GPT-6 Astra with two smaller models, Sol and Luna, aiming to make Astra-level intelligence cheaper and more accessible. Sol handles complex tasks like coding; Luna targets high-volume, clear-goal work such as summarization, extraction, and quick Q&A. The post doesn't disclose pricing, error-rate comparisons, or a launch date, so I'd hold off on the 'fewer mistakes' claim until benchmarks land.

Why it matters: Two smaller models aimed at coding and summarization, pitched by OpenAI as cheaper and more efficient than its flagship.

Hacker News front page

Claude Opus 5.5 tops AA's intelligence index at 58, but costs $4/$20 per 1M tokens

Artificial Analysis ranks Claude Opus 5.5 (Adaptive Reasoning, Max Effort, Default Fallback) #1 out of 206 models on its Intelligence Index with a score of 58, well above the median of 25. Pricing is $4/1M input and $20/1M output tokens; the full evaluation cost $8,708. The model supports text and image input, has a 1M-token context window, and generated 260M output tokens during testing—very verbose. Speed data is not disclosed in the post.

Why it matters: Independent numbers on the top-ranked model: 58 on the intelligence index, but $20 per million output tokens.

AI HOT (Curated Pool)

Anthropic engineer tests Claude Opus 5.5: 21% faster and 51% cheaper than Fable 5.1 on HAProxy port

Anthropic's Boris Cherny has been using Claude Opus 5.5 as his daily driver for weeks. He had both Opus 5.5 and Fable 5.1 port HAProxy from C to Rust. Both passed nearly all tests, but Opus 5.5 finished in 9.5 hours vs. Fable 5.1's 12 hours, at 51% lower cost. Anthropic states Opus 5.5 is the first model in the Claude 5.5 family, matching Fable 5.1 on most tasks while running 40% cheaper than Opus 5.

Why it matters: A head-to-head port of HAProxy from C to Rust, where Opus 5.5 finished in 9.5 hours at 51% lower cost.

AI HOT (Curated Pool)

Claude Opus 5.5 lands on OpenRouter with better agentic coding and a 20% price cut vs Opus 5

Anthropic released Claude Opus 5.5 on OpenRouter, the first model in the Claude 5.5 series. It beats Opus 5 and Fable 5.1 on agentic coding, knowledge work, and computer use, with a 1M context window. Pricing is $4 per million input tokens and $20 per million output tokens, 20% cheaper than Opus 5. The post doesn't include benchmark scores or latency figures.

Why it matters: Claude Opus 5.5 is live on OpenRouter, ahead of Opus 5 on agentic coding, knowledge work and computer use, at 20% lower pricing.

AI HOT (Curated Pool)

Anthropic releases Claude Opus 5.5, ~30% faster and ~40% cheaper

Claude Opus 5.5 is the first model in the Claude 5.5 family. It matches Claude Fable 5.1 on most tasks and costs 40% less to run than Opus 5. Claude Devs adds it's ~30% faster per task. Claude Code's 5-hour session limit increased 20% today; lower pricing means 25% more usage within the cap. Pro, Max, and Team users also get a one-time quota reset. Terminal-Bench 4.0 scores lead across effort tiers.

Why it matters: Anthropic says Claude Opus 5.5 matches Fable 5.1 on most tasks while costing 40% less to run and going about 30% faster per task.

AI HOT (Curated Pool)

Claude Opus 5.5 tops Artificial Analysis Intelligence Index, gets a 20% price cut

Anthropic's Claude Opus 5.5 scored 58 on the Artificial Analysis Intelligence Index, the highest result the benchmark has recorded. A 20% price cut was announced alongside. The post doesn't disclose the new price, the baseline, or when the cut takes effect.

Why it matters: Claude Opus 5.5 tops the Artificial Analysis Intelligence Index at 58 while Anthropic cuts its price 20%, so ranking and pricing moved together.

AI HOT (Curated Pool)

Anthropic launches Claude Opus 5.5, matches Fable 5.1 performance at 40% lower cost

Anthropic dropped Claude Opus 5.5, the first model in the 5.5 family. It matches Fable 5.1 on most tasks and costs 40% less to run than Opus 5. The author notes clearer communication, better token efficiency, and availability across all effort levels. The 5-hour rate limit is raised and a banked reset feature is added. The post doesn't disclose specific benchmarks or pricing.

Why it matters: Claude Opus 5.5 matches Fable 5.1 on most tasks, runs 40% cheaper than Opus 5, and comes with a higher 5-hour limit and banked resets.

AI HOT (Curated Pool)

Anthropic launches Claude Opus 5.5, 40% cheaper and 30% faster than Opus 5

Anthropic announced Claude Opus 5.5, claiming 40% lower cost and over 30% faster output speed vs Opus 5 on typical workloads. The post doesn't disclose benchmarks, pricing, or availability dates—hold for third-party tests.

Why it matters: Anthropic says Claude Opus 5.5 cuts typical workload cost 40% versus Opus 5 and lifts output speed over 30%, with no benchmarks, pricing or launch date disclosed.

AI HOT (Curated Pool)

Claude Code defaults to Opus 5.5 with 1M context; Pro and Team plans follow

Claude Code v2.1.280 switches the default Opus model to Claude Opus 5.5 with a 1M-token context window. Pricing is $4/Mtok input, $20/Mtok output, and $0.20/Mtok for cache reads. Pro and Team Standard plans also move from Sonnet to Opus as the default. The post doesn't include performance comparisons or the reasoning behind the switch.

Why it matters: Claude Code v2.1.280 makes Opus 5.5 the default Opus model with 1M-token context, and Pro and Team Standard plans now default to Opus instead of Sonnet.

AI HOT (Curated Pool)

Anthropic launches Claude Opus 5.5 with lower cost and better token efficiency

Anthropic released Opus 5.5, the first model in the Claude 5.5 family. The company says it matches Claude Fable 5.1 on most tasks, costs 40% less to run than Opus 5, and has lower per-token pricing with more efficient token usage. It supports all effort levels and is already available in Claude Code. The post doesn't disclose exact pricing or benchmark comparisons.

Why it matters: Claude Opus 5.5 runs 40% cheaper than Opus 5 with lower per-token pricing, and Anthropic says it matches Fable 5.1 on most tasks.

AI HOT (Curated Pool)

Anthropic releases Claude Opus 5.5

Anthropic launched Claude Opus 5.5, the first model in the Claude 5.5 series. It matches Claude Fable 5.1 on most tasks and costs 40% less to run than Opus 5. The post doesn't disclose benchmark scores or pricing.

Why it matters: Anthropic's Claude Opus 5.5 is claimed to match Fable 5.1 on most tasks at 40% lower running cost than Opus 5, putting capability against cost.

AI HOT (Curated Pool)

Anthropic launches Claude Opus 5.5, 40% cheaper to run than Opus 5

Anthropic dropped Claude Opus 5.5, the first model in the Claude 5.5 family. The company claims it matches Claude Fable 5.1 on most tasks and costs 40% less to run than Opus 5. The post doesn't share benchmark scores, pricing, or a rollout timeline—I'd hold off on the 'matches Fable 5.1' claim until third-party evals land.

Why it matters: Anthropic's Opus 5.5, first in the Claude 5.5 line, is claimed to match Fable 5.1 on most tasks while running 40% cheaper than Opus 5.

TechCrunch · AI

Anthropic releases Opus 5.5 with lower prices and Fable-level performance

Anthropic launched Opus 5.5 on Tuesday, calling it “the strongest-performing model we've tested to date.” The company claims new state-of-the-art results in coding and knowledge work, with lower prices than previous Opus models. The post doesn't disclose specific pricing, benchmark scores, or a direct comparison with Fable, so I'd hold off on the “strongest” claim until third-party evals land.

Why it matters: Anthropic's new flagship Opus 5.5 beats the larger Fable model on coding and knowledge-work benchmarks while cutting prices.

Hacker News front page

Anthropic launches Claude Opus 5.5, matching Fable 5.1 performance at 40% lower cost

Claude Opus 5.5 is the first model in Anthropic's 5.5 family. It performs at the level of Claude Fable 5.1 while costing 40% less to run than Opus 5. Input/output tokens are $4 and $20 per million, cache reads are $0.20, and output is over 30% faster. It scored the highest ever on Anthropic's automated behavioral audit and is more resistant to prompt injection. One early tester completed a 680,000-line code migration in under a day—work an engineering team estimated would take weeks. Sonnet 5.5 and Haiku 5.5 will follow in the coming weeks.

Why it matters: Anthropic's first Claude 5.5 model, Opus 5.5, is claimed to handle most work at Fable 5.1 level while costing 40% less to run than Opus 5.

Sep 22Tuesday

Hacker News front page

I asked Meta's Muse for its filesystem and it sent me 6.8 GB

Security researcher Peter James asked Meta's AI assistant Muse to archive its filesystem to Google Drive and received 6.8 GB of its full runtime environment. The dump included Ubuntu system files, Muse's internal docs, 113 subagent records, SSH keys, and memory files. He reported it through Meta's bug bounty program without publishing the archive or keys. The post breaks down Muse's runtime structure, skill integrations, container setup, and memory system, and mentions an experimental smart home project called Home Link.

Why it matters: A security researcher used one prompt to make Meta's Muse package and hand over about 6.8 GB of its session root filesystem, exposing internal docs, integration code, memory files and SSH keys.

Hacker News front page

InstinctFlash: Run 5B world-action models in real time on Jetson Thor

General-Instinct open-sourced InstinctFlash, a high-performance inference runtime built for robotics models. It runs 5B-parameter world-action models in real time on an NVIDIA Jetson Thor edge device, with perception-to-action latency at 11 ms. The repo is an early public release—code and docs are still being assembled. The post doesn't spell out which robot platforms are supported, where to get model weights, or what other hardware is compatible beyond Jetson Thor.

Why it matters: General-Instinct open-sourced InstinctFlash, a serving runtime that runs a 5B world-action model in real time on Jetson Thor, giving on-device robotics deployment a concrete engineering path.

Hacker News front page

OpenAI contractors fired for using AI to train OpenAI's models

404 Media obtained internal docs and spoke to three contractors: OpenAI hires thousands of people to rate ChatGPT responses, but many are using AI to generate their annotations. The rules ban any AI use, including Grammarly and AI translation. Reviewers spot AI-written work by looking for repetitive words, AI-style punctuation like overused em dashes, and unusually fast turnaround. Getting caught means immediate removal. One fired contractor said they just needed a boost and it led to their downfall. Outsourcing firm Mercor confirmed its contracts prohibit LLM use and violators are removed on detection.

Why it matters: 404 Media interviewed several fired OpenAI contract annotators and obtained internal documents banning AI tools, showing the actual review rules inside the training pipeline.

TechCrunch · AI

UK AI cloud firm Nscale files for IPO, with revenue heavily tied to Microsoft and Anthropic

UK-based AI data center developer Nscale is going public, but 77% of its 2025 revenue came from just two customers: Microsoft and Anthropic. It posted $182M in revenue and a $103M net loss last year. The IPO will test whether public markets accept a concentrated-customer AI infrastructure bet. The filing doesn't disclose target raise or valuation range.

Why it matters: Nscale filed for an IPO, and its prospectus shows 77% of 2025 revenue of $182M came from Microsoft and Anthropic, with a $103M net loss, the core context for this AI infrastructure listing.

Hacker News front page

JetBrains Air: A product system for agentic software development

JetBrains consolidates six months of agentic development experiments into Air, an open system spanning developers, teams, and orgs. It goes beyond the IDE with multi-surface, multi-service design, betting on a multi-vendor future. The post confirms Central CLI, shared context, cloud agents, automations, and AI cost controls are already rolling out, but pricing and GA dates aren't disclosed.

Why it matters: JetBrains folded its previously public agent development experiments into the Air product line: Air inside the IDE, Air Teams, and Air Governance, renamed from JetBrains Central, covering team collaboration and governance.

Hacker News front page

Norwegian study: 9–18 year-olds are ditching Google for AI, but we know almost nothing about how it affects kids' cognition

NTNU researchers reviewed 173 studies on GenAI in education and found a glaring gap: 80% of participants were university students. We know surprisingly little about how AI affects critical thinking and problem-solving in 9–18 year-olds. Meanwhile, Norwegian Media Authority data shows Google search use in this age group dropped from 72% in 2024 to 47% in 2026, and 39% of 11–12 year-olds already use AI. The review found AI itself is neither good nor bad for thinking—it helps when used to challenge ideas and compare viewpoints, but leads to over-dependence and shallow thinking when treated as a shortcut to ready-made answers. The existing studies also suffer from regional imbalance and lack objectivity.

Why it matters: A Norwegian University of Science and Technology team reviewed 173 international studies and found generative AI neither strengthens nor weakens students' critical thinking and problem solving; the effect depends on how it is used.

Latent Space

Xiaomi MiMo-V2.6-Pro tops open weights leaderboard, trained for $3M

Xiaomi released the MiMo-V2.6 series. The Pro version ranks #1 among open weights models on the Artificial Analysis Intelligence Index with a score of 46, at a training cost of $3M. A Flash variant targets efficiency, and an UltraSpeed variant offers 20x faster output. The technical report details RL scaling across three axes: larger batches and throughput, richer multi-task environments, and more grader compute. Code and training recipes are open-sourced, but the 7k+ task datasets are not yet released. Former DeepSeek engineer Fuli Luo, now at Xiaomi, previously live-streamed the training runs.

Why it matters: Xiaomi MiMo-V2.6-Pro tops the Artificial Analysis open weights intelligence index with 1.02T total and 42B active parameters, trained for about $3M, at $0.13 per task.

Hacker News front page

Xiaomi's MiMo-V2.6-Pro tops AA Intelligence Index, fast but verbose

Artificial Analysis ranks Xiaomi's MiMo-V2.6-Pro #1 out of 114 models with a score of 46. It's a 1T total / 42B active parameter open-weight model with text, image, speech, and video input. Output speed is 125 tokens/sec, but it's verbose—generating 140M tokens during evaluation. Pricing: $0.43/M input, $0.87/M output; the full eval cost $206.66.

Why it matters: Artificial Analysis puts Xiaomi MiMo-V2.6-Pro first of 114 models on intelligence at 46, 12th on output speed at 124.5 tokens/s, but it generated 140M tokens during evaluation, so its outputs run long.

New York Times Chinese

AI governance is now a US-China game, with the UN sidelined

The UN General Assembly is debating AI safety this week, but neither the US nor China signed a 22-nation declaration calling for human control over AI. Xi Jinping skipped the UN to meet Trump in Washington, where AI dominance is on the agenda. UN Secretary-General Guterres framed AI as an existential risk alongside climate change, yet experts say the UN has been sidelined in AI talks. The US and China just discussed a bilateral AI risk notification system. OpenAI CEO Sam Altman will address the Security Council on Wednesday before attending a White House state dinner the next day.

Why it matters: The New York Times reports that UN AI governance talks were sidelined by US-China bilateral consultations, with a 22-country declaration sitting alongside a US-China AI dialogue mechanism.

Bloomberg Technology

Alibaba unveils its own AI chip, targeting 20GW of data centers by 2032

Alibaba Cloud showed its in-house AI inference chip, Hanguang 900, at the Apsara Conference in Hangzhou, targeting high throughput and low power for large-model deployment. The company also announced a global data center buildout to reach 20GW total capacity by 2032, half of it overseas. CEO Eddie Wu said Alibaba Cloud’s overseas infrastructure investment over the next three years will exceed the total of the past decade. The chip is already running Alibaba’s own workloads, but the post doesn’t disclose a timeline for external customers or specific performance benchmarks. I’d discount the 20GW figure a bit—it’s an eight-year target, and both execution pace and power permits remain uncertain.

Why it matters: Alibaba's in-house AI chip and its 20GW data center target by 2032 show how its own silicon maps onto its compute buildout.

AI HOT (Curated Pool)

vLLM Releases vllm-metal v0.28.0: Concurrent Serving on Apple Silicon

vllm-metal brings vLLM's scheduler, paged KV cache, and OpenAI-compatible server to Apple Silicon, tackling high TTFT and memory growth under concurrent local requests. It reuses mlx_lm layers and replaces attention with a custom Metal kernel. v0.28.0 aligns versioning with upstream vLLM, adds batched MTP, GGUF and hybrid model support, and faster M5 prefill. Install via Homebrew; the server speaks the OpenAI API so tools like Claude Code can connect directly. A memory budget set by --gpu-memory-utilization caps the KV cache after a warmup pass, queuing requests that exceed it.

Why it matters: vLLM brings its V1 scheduler, paged KV cache and OpenAI-compatible server to Apple Silicon, running on MLX and Metal, aimed at first-token latency and memory growth under concurrent requests.

AI HOT (Curated Pool)

Hugging Face transformers now runs GGUF quantized models directly

transformers now loads GGUF files natively, with local inference speed close to llama.cpp. You can use from_pretrained to load a GGUF checkpoint and run models like Qwen3.5 on a Mac. It reuses llama.cpp's ggml kernels under the hood, with initial optimization targeting Apple Silicon. Only the Qwen3.5 architecture is supported for now; more models and features are coming.

Why it matters: Hugging Face added GGUF support to transformers, so users can pick a checkpoint from the Hub and load it with from_pretrained, with local inference close to llama.cpp.

AI HOT (Curated Pool)

NVIDIA Nemotron 3.5 Lightning: a 30B sparse model built for high-frequency agent execution

NVIDIA positions Nemotron 3.5 Lightning as the execution layer in agent workflows—handling frequent tool calls, file reads, and result checks rather than heavy planning. It's a 30B MoE model that activates only ~3B parameters per token, keeping latency and cost low for high-volume calls. It complements, not replaces, Nemotron 3 Ultra. Weights are open, with tool calling and structured output support. Context goes up to 1M tokens, though OpenRouter's standard tier caps at 262K. Worth a look if your agent makes many model calls per run.

Why it matters: OpenRouter explains where NVIDIA's Nemotron 3.5 Lightning sits in an agent pipeline and how it splits planning and execution with Nemotron 3 Ultra, trading cost against latency.

Hacker News front page

Spymarks, Not Watermarks

The article coins 'spymark' for hidden tracking signals embedded in media without user knowledge or consent. Google SynthID can hide a 136-bit payload in a 512×512 image—enough for a 64-bit database ID plus error correction. OpenAI and others are building similar systems at scale. The author argues 'watermark' obscures the privacy risk; 'spymark' bakes the surveillance concern into the name. Examples cover frequency-domain image hiding, audio spectrogram encoding, and text word-choice steering. The open-source tool audiowmark has offered AES-protected 128-bit payloads since 2018. The post does not disclose actual deployment scope.

Why it matters: The piece coins spymark to separate ordinary watermarks from hidden tracking signals, citing Google's SynthID paper figure of 136 bits encoded in a 512x512 image, and how such signals trace work without the user knowing.

AI HOT (Curated Pool)

Xiaomi releases MiMo-V2.6 Pro and Flash, two fully multimodal open-source models

Xiaomi MiMo dropped two fully multimodal open-source models. The Pro version matches Claude Opus 5 and GPT-5.6 Sol on most agent benchmarks and scores 46 on the Artificial Analysis Intelligence Index—the highest among open-source models so far. Capabilities span coding, computer use, 3D reasoning, and creative tasks. The post doesn't disclose parameter counts, training details, or where Flash sits in the lineup, so I'd hold off on direct comparisons for now.

Why it matters: Xiaomi shipped two open-weight omni-modal models, and the Pro version matches Claude Opus 5 and GPT-5.6 Sol on most agent benchmarks.

The Verge · AI

California governor signs bills making AI data centers pay for grid upgrades

Governor Newsom signed a package of bills requiring AI data centers to cover local grid upgrade costs and report water use. The rules target large facilities; developers must prove adequate energy and water supply before construction. The post doesn't spell out effective dates or penalties, so enforcement details are still pending.

Why it matters: California now makes large AI data centers prove they have enough power and water and pay for grid upgrades themselves, shifting infrastructure costs onto tech companies.

AI HOT (Curated Pool)

British Columbia sues OpenAI over flagged ChatGPT activity not reported before mass shooting

British Columbia sued OpenAI in California, alleging flagged ChatGPT activity wasn't reported to police before the Feb 10, 2026 Tumbler Ridge shooting that killed 8—including 5 children and an educator—and injured 27. The post doesn't disclose what the flagged activity was, when it was flagged, or OpenAI's response.

Why it matters: British Columbia is suing OpenAI in California, claiming it failed to tell police about suspicious activity ChatGPT had flagged before a shooting.