IBM trims tokens 59%, Alpert: own every LLM line, Datasette ships swap API

Sources

Thinking of ACE? We Can Do It with Fewer Tokens huggingface.co

There are no lossless transformations of natural-language text simonwillison.net

There are no lossless transformations of natural-language text Sophie Alpert shares her “internal policy on acceptable use of AI writing by engineers”. It’s a short read (supporting its own recommendations) and really good. If you chose to have LLMs help massage your writing the following rule seems crucial to me: You must stand behind every idea and every sentence in your docs . It is your responsibility to make sure that the entire document is representative of your own thoughts before you sh…

datasette-upload-dbs 0.5a0 simonwillison.net

Release: datasette-upload-dbs 0.5a0 This plugin has been around for a while - it lets users upload a brand new SQLite database to a hosted Datasette instance, at which point that database will start being served by that instance. It can also be used to atomically swap a database with a more recent version. The uploaded database is saved to a file, verified, then swapped in so /name starts serving the new one. The new release adds a formalized API, so you can replace an existing database (or add…

References

SambaNova (ACE authors’ blog) sambanova.ai

ACE consistently outperformed strong baselines by +10.6% on agentic tasks… matched top production-level agents like IBM-CUGA (powered by GPT-4.1) on the hardest test-challenge split using a smaller open-source model (DeepSeek-V3.1).

ACE paper (arXiv:2510.04618) arxiv.org

Structured, incremental delta updates preserve detailed procedural knowledge… rather than collapsing history into brief summaries.

Moogician blog — Trustworthy Benchmarks moogician.github.io

24.4% of certain successful trajectories on SWE-bench merely used git log to copy answers from commit history… agents could achieve near-perfect scores by exploiting environment configurations rather than solving problems through reasoning.

Graphlit — Survey of AI Agent Memory Frameworks graphlit.com

Mem0’s hybrid vector-and-graph approach reportedly achieves ~92.5% on LoCoMo versus 52.9% for native OpenAI Memory… MemGPT/Letta remains a favorite despite high token cost.

IBM announcement — ALTK-Evolve ibm.com

Full mode uses an MCP server with Milvus/PostgreSQL vector stores; Lite mode targets rapid setup with Claude Code and IBM Bob. Includes a Policy Guardrail component that checks agent outputs against organizational compliance rules.

Evermind — MemGPT alternatives evermind.ai

MemGPT relies on the agent to actively edit its own memory blocks via tool calls, often leading to increased latency and ‘stuck thinking’ loops where the model fails to execute the memory tool.

The Next Web on HBR ‘workslop’ study thenextweb.com

41% [of workers] had received workslop in just the previous month… each instance of workslop required nearly two hours of manual rework to resolve… the ‘hidden tax’ of these incidents at approximately $186 per employee per month.

BetterUp Labs / Stanford Social Media Lab workslop report betterup.com

Approximately 53% of workers… reported that receiving workslop damaged their trust in their colleagues… colleagues who rely too heavily on unedited AI drafts are perceived as less creative, less capable, and more disengaged.

The State of Brand — coverage of Clay’s policy thestateofbrand.com

Leadership, including COO Varun Anand, distinguishes between the product—which focuses on research, signal assembly, and lead enrichment—and the internal communication of the ‘stated view’ of what those signals mean.

Stanford HAI — AI detectors biased against non-native English writers hai.stanford.edu

Seven popular detectors misclassified over 61% of TOEFL essays as AI-authored, while native-speaker essays were identified correctly nearly 100% of the time.

Bluesky discussion of Alpert essay bsky.app

Cryptographer Filippo Valsorda has expressed a ‘nuclear opinion’ regarding the burden of review: if an author uses an LLM to generate content, it may be inefficient or even insulting to require a human to review it until another model has first verified it for basic quality.

Sateesh Nori, ‘Don’t use AI to write unless…’ (Substack) sateeshnori.substack.com

Bypassing the ‘struggle’ of word choice and sentence structure effectively removes the recursive feedback loop between writing and thinking… over-reliance on AI is thus viewed as a ‘crutch’ that may produce better grades in the short term but ultimately results in ‘cognitive debt.’

datasette.cloud blog — datasette-litestream datasette.cloud

datasette-litestream allows a Datasette instance to replicate its SQLite database to S3-compatible storage, enabling point-in-time recovery and persistence across restarts… turning Datasette into a full-fledged CRUD framework where changes are persistent and recoverable, even on ephemeral hosting like Fly.io or Cloud Run.

litestream.io/alternatives litestream.io

LiteFS is a FUSE-based file system that replicates SQLite across a cluster of machines… rqlite and dqlite use Raft-based consensus to ensure every write is committed across a quorum of nodes, trading some write throughput for higher durability guarantees compared to Litestream’s asynchronous one-second backup window.

Datasette docs (datasette.io) docs.datasette.io

The plugin utilizes Datasette’s internal remove_database(name) and add_database(db, name=name) methods… db.close() must be called explicitly before remove_database() is invoked to clear the thread-local connection cache, otherwise Datasette continues serving data from an old version of the file even after it has been replaced on disk.

Simon Willison — 1.0a38 security note (contextual) simonwillison.net

On August 6, 2026, Datasette 1.0a38 addressed a SQL injection vulnerability that affected instances serving mixed public and private tables within the same database… administrators were urged to disable raw SQL query permissions if they managed sensitive data.