NEWSFERENCE
TUE · NOV 17 · 2026 · 19:16 AWST
LIVE
$ today --liveF1TodayF2YesterdayF3ArchiveF4About
NEXT SCAN 19:30

SUN · OCT 18 · 2026

45 clusters from 78 articles.

TOP CLUSTER · TIER 1
Anthropic workforce matures as junior headcount shrinks and meeting time grows.

Anthropic's workforce composition and the nature of the work being done have shifted significantly over the past year, with the company becoming less junior overall and employees spending more time in meetings.

ACROSS THE WIRE · TIER 3
20 ROWS
#
ENTITY
HEADLINE
SRC
X
TIME
10
MISC
Openspline launches open-source realtime talking avatar generator for voice AI agents.
1
—
30d
11
DEEPSK
Simon Willison releases web-edit 0.1a2, demos with DeepSeek 4.1 Flash 1017.
1
—
30d
12
MISC
AnimLang tool enables animation creation via LLM workflows using GSAP and ThreeJS
1
—
30d
13
QWEN
Community runs Qwen 3.8 27B at 262k context on single 16GB AMD GPU via offload.
1
—
30d
14
MISC
Community runs Fish Audio Drama 3 on 16GB AMD GPU via RAM offload.
1
—
31d
15
DEEPSK
Xolotl inference engine runs 552B DeepSeek-V4.1-Flash on single RTX 4090 via RAM offload.
1
—
31d
16
DEEPSK
DeepSeek V4.1 Flash 1017 benchmarks show 57 tok/s text on 2x DGX Spark.
1
—
30d
17
QWEN
Basalt enables NVFP4 quantization for Qwen 3.8 Flash-Next on Blackwell.
1
—
30d
18
DEEPSK
DeepSeek v4.1 Flash 1017 vision tower ported to Dwarfstar ds4 with 2-bit experts.
1
—
30d
19
AMZN
Amazon deploys AI surveillance to monitor union risk in UK warehouses
Policy
1
—
31d
20
META
OnlyFans creators challenge Meta over Muse AI sexual content permissions.
Policy
1
—
30d
21
MISC
Opencode security breach prompts key rotation advisory.
Policy
1
463
31d
22
META
Meta removes web scraping language that raised AI training concerns.
Policy
1
—
31d
23
ANTHRP
Anthropic consults law firms on AI agent negotiation norms.
Policy
1
7.2k
30d
24
ANTHRP
Anthropic proposes AI safety bill tied to 50 deaths or $1B damages threshold.
Policy
1
—
30d
25
GOOGL
German courts hold Google liable for AI overview search results.
Policy
1
—
30d
26
LINUX
Linus Torvalds calls anti-AI regulations idiotic
Policy
1
—
31d
27
MISC
Community benchmarks nine AI models on single 16GB AMD GPU for coding
Research
1
—
30d
28
MISC
Harvard and Cambridge join Hong Kong forum on humanities in AI era
Research
1
—
30d
29
MISC
Community debates older models for technical writing amid new model complaints
Research
1
—
30d