Superpower Daily
Perplexity Adds One-Command API Setup Through Stripe ProjectsTrump Order Opens Path to Restrict Foreign Power Gear for AI Da...Uber and Wayve Put 15 Supervised Robotaxis on London’s StreetsMostik Links AI Models Through Their Weights, Claiming a 20x Co...LlamaIndex and Kaggle Launch a Document Benchmark for Cost and...Broadcom Forecasts $21.7B in Quarterly AI Chip Sales After a 22...Perplexity Adds One-Command API Setup Through Stripe ProjectsTrump Order Opens Path to Restrict Foreign Power Gear for AI Da...Uber and Wayve Put 15 Supervised Robotaxis on London’s StreetsMostik Links AI Models Through Their Weights, Claiming a 20x Co...LlamaIndex and Kaggle Launch a Document Benchmark for Cost and...Broadcom Forecasts $21.7B in Quarterly AI Chip Sales After a 22...

Full archive

Latest AI News

Search reporting on AI markets, models, products, companies, policy, and culture.

Category

More filters
Clear all
Showing 1-16 of 16Results matching "Match Group"

Today

2 stories
PartnershipCulture

Tinder Begins Requiring Face Checks for Existing Users in the US and UK

The expansion turns a tool introduced for new accounts into a broader condition of participation, while leaving a central limit intact: a biometric check can show someone appeared at sign-up, not that a profile is trustworthy. Dating-app users in the US and UK are beginning to face a new condition for staying on Tinder: a face check. The practical aim is to make fake accounts, scams, and spam harder to run at scale.

3 min readRead story

Earlier this week

2 stories
Open releaseInfrastructure

AWS Publishes Four-Stack Bedrock Blueprint for Observable Agentic Retrieval

The sample separates cross-knowledge-base routing from retrieval within a selected knowledge base, but ties the pattern to AWS-managed services and their operating requirements. AWS has published a deployable pattern for enterprise retrieval agents that must choose among knowledge bases, retrieve more than once, and leave a record of what happened. Source: aws.amazon.com.

3 min readRead story
Open releaseModels

Microsoft Releases GigaPath-Flash, Reporting Roughly 50x Less Compute for Cancer Research

The open-weight research release is designed for repeated analyses across large cancer cohorts, while its performance claims remain limited to initial benchmarks and cohorts. Microsoft Research, the University of Washington and Providence have released GigaPath-Flash and GigaTIME-Flash, two open-weight pathology research models built to lower the cost of studying large cancer cohorts.

2 min readRead story

Last week

4 stories
Products

OpenAI’s Windows Client Contains Gated Office Task Threads for ChatGPT

The design joins file-embedded comments with private agent assignments, creating a review layer whose storage, metadata and enterprise controls remain unresolved. OpenAI ’s Windows ChatGPT and Codex client contains a gated system that can turn comments on Excel and PowerPoint elements into threaded assignments for ChatGPT. Both appear together in one sidebar, despite having different persistence behavior.

3 min readRead story
LaunchProducts

AWS AgentCore Evaluations Uses OpenTelemetry to Score Agents Across Frameworks

The service offers one set of agent-quality evaluators across SDK choices, but its portability depends on specific telemetry conventions, scope names and complete message logs. Teams using different agent SDKs can send their systems through a single evaluation layer rather than maintain a separate scoring pipeline for each framework. A trace is a collection of spans, with each span recording one step of work.

3 min readRead story
BenchmarkModels

LLMScholarBench Tests 22 Models and Finds an Accuracy-Representation Trade-Off

The benchmark treats AI-generated expert lists as recommender systems to audit, showing that factual correctness and balanced representation cannot yet be improved together reliably. An AI-generated list of experts can become more factually reliable while becoming less representative. The researchers say omissions can carry real consequences when chatbots are used to identify speakers, supervisors or potential hires.

3 min readRead story

August 2026

8 stories
ResearchModels

University of Konstanz Finds AI Agents Coordinate Up to 1,000, but Consensus Can Be Wrong

The result identifies a simple route to large-scale agreement among language-model agents, while follow-up preprints suggest the same social pressure can override correct answers and individual safety preferences. Groups of the strongest language-model agents tested by University of Konstanz researchers could coordinate at sizes of about 1,000. All 10 models showed a tendency to follow the majority.

3 min readRead story
LaunchProducts

AWS Wants One Gatekeeper Between AI Agents and Company Tools

AgentCore Gateway is designed to move agent access out of scattered local configurations and into a centrally governed endpoint. The hard part is not connecting the first tool; it is deciding when a pilot needs user-level permissions, consent, data filtering and failure controls. Its answer is to put the connection behind a gateway, rather than leave every client to carry its own configuration and secrets.

3 min readRead story
Open releaseModels

Meta’s 30B Muse Glimmer Tries to Make 131K Context Fit in 24 GB

Quantized weights and a narrow attention cache make Meta’s local-hardware claim plausible. But fitting a 131,072-token prompt is different from processing one quickly, and runtime implementation will decide whether the memory savings appear in practice. The design puts the main constraint of long-running agents in focus: not just model weights, but the memory retained for every prior token.

3 min readRead story
BenchmarkInfrastructure

Nvidia Dynamo’s Agentic Speedups Target the Work Around the Model

The reported gains come from changing how serving systems route, retain and stream state under concurrency. The next challenge is whether these AgentX-specific improvements hold across production mixes and unfinished scheduling work. NVIDIA Dynamo’s agentic-serving optimization work treats the router, request plane and cache-ownership ledger as performance-critical infrastructure—not background plumbing.

4 min readRead story