AI & Machine Learning

Synthetic Sincerity review – Marc Isaacs’ AI interrogation grapples with identity and existence

A combination of fact and fiction leaves the celebrated documentarian’s puzzling project about software training wanting for depth Marc Isaacs’ new film is a curious, intriguing, semi-sincere affair that I couldn’t make friends with. It is an odd, shallow piece of work about artificial intelligence that is itself exasperatingly artificial, a self-aware docudrama hybrid. Isaacs is, or rather pretends to be, licensing the vivid characters from his previous, acclaimed documentaries to a fictional AI research lab called Synthetic Sincerity at the fictional University of Southern England, so that the lab’s software can be “trained” in the creation of AI human figures on screen. The lab’s research staff are played by actors, or at any rate people acting; these include Lebanese independent film-maker Lynn El Safah. Isaacs has amusing scripted conversations about this project with a disapproving AI avatar on screen, like Max Headroom of old, whose face is digitally modelled on Romanian actor Ilinca Manolache, from Radu Jude’s Do Not Expect Too Much from the End of the World. The film, however, does not show the process by which Manolache was approached and her face transformed into an AI figure.

Did AI Help With the Work? Saying Yes Could Mean Missing Out on a Raise or Promotion.

Managers are pushing employees to use AI, but adopting it could mean negative consequences for workers.

Why Performance per Watt Is the Ultimate Metric for AI Infrastructure Efficiency

Power is AI infrastructure’s inescapable constraint. How many tokens an AI factory can generate within a fixed power budget determines its revenue and profitability. Because of this, performance per watt — a metric that can’t be gamed, only earned through real-world results — is the foundation for AI factories. As agentic AI drives token demand

As AI Platforms Move Away From Unlimited, Sogni AI Launches a $20 Fair-Use Unlimited Plan on Community GPUs

As AI Platforms Move Away From Unlimited, Sogni AI Launches a $20 Fair-Use Unlimited Plan on Community GPUs

Surprise: Many People Prefer AI Audiobooks, New Study Finds

Even with some of the backlash surrounding the AI Michael Caine-led audiobook of Homer’s “Odyssey” poem, it appears many listeners prefer an artificially generated voice to a human one. A new Edison Research at SSRS study commissioned by Spoken, an AI audiobook company, revealed that the company’s multi-voice productions achieved a higher favorability rating than human-narrated

Exclusive: Google's Hassabis calls for U.S.-led global AI watchdog

Demis Hassabis, Google DeepMind co-founder and CEO, is calling on the U.S. to establish a new AI watchdog with the power to screen the world's most advanced models — and coordinate an industry-wide slowdown if dangers mount. Hassabis, the Nobel laureate behind Gemini, lays out the plan in a personal manifesto published Tuesday morning, "A Framework for Frontier AI and the Dawning of a New Age." Why it matters: In an exclusive interview with Axios, Hassabis said the time has come for a more "systematic" approach to AI regulation — funded by the industry, staffed by world-class technical experts, and answerable to the U.S. government. Today's AI-driven cyber risks are "warning shots," Hassabis told us from his London base. Within 18 months, he said, those capabilities — plus far graver biological and nuclear threats — could live inside open-source models beyond any government's control. Hassabis emphasized to us that risks will come from the major labs' more powerful future proprietary models, not just open-source models. "What we collectively do now," he writes in his manifesto, "will determine how the next phase of civilization unfolds." Behind the scenes: Hassabis has spent months quietly building support for the plan, briefing the Trump administration, fellow lab leaders and European officials before going public. "The noises I've been hearing are very positive," he said of his talks with the administration, which had embraced a laissez-faire approach to AI regulation prior to the Mythos scare. Hassabis, a scientist who commands rare respect across AI's warring camps, says the other major lab leaders agree at a high level: "This is where the industry needs to go." His timeline is aggressive. "Months," Hassabis said, ideally with the new body operational "before year-end." How it works: Hassabis is proposing an AI standards body modeled on FINRA (Financial Industry Regulatory Authority), the private, industry-funded watchdog that polices Wall Street under SEC overs

OmniRoute: The Open-Source AI Gateway Slashing Token Costs by 95%

Optimizing LLM Costs: Meet OmniRoute AI coding agents are boosting developer productivity, but they consume a massive amount of tokens. For developers running coding assistants like Claude Code, Cursor, or Copilot all day, API subscription fees and token costs can scale quickly. OmniRoute is an open-source AI gateway designed to tackle this problem. By combining advanced token compression with smart routing, it provides a single unified gateway connecting your local tools to over 250 AI providers—including 90+ free endpoints. OmniRoute acts as a local proxy or containerized gateway. Instead of pointing your coding tools directly to OpenAI or Anthropic APIs, you route them through OmniRoute. The gateway manages API keys, balances traffic, compresses prompts, and routes requests to the cheapest or most reliable provider. The standout feature of OmniRoute is its custom token compression. By optimizing prompt formats and stripping redundant context, the stacked compression engine saves between 15% and 95% of token usage per request. This dramatically lowers API costs for developers working on large codebases. OmniRoute links to over 250 API endpoints, giving developers access to a huge range of LLM models. It supports 90+ permanently free tiers, allowing you to run dev agents without paying premium fees. If an API provider goes down, hits a rate limit, or experiences high latency, OmniRoute handles it gracefully. It automatically falls back to alternative models or providers in real-time, keeping your coding session running smoothly without crashes. Designed for the modern agentic ecosystem, OmniRoute natively supports MCP and Agent-to-Agent (A2A) communication, making it highly compatible with advanced coding agents. OmniRoute can be deployed on your local machine using Docker or run as a Desktop app/PWA. Simply spin up the Docker container, input your preferred API keys (or select free providers), and update the API endpoint URL in your IDE extensions to point to your

George Lucas Goes To The Dark Side, Says AI Is ‘The Future’

The Star Wars creator compares not using AI to riding a horse in the face of cars

StepFun Unveils the First Agentic AI Phone Ahead of Apple and OpenAI

Tencent-backed StepFun unveiled the StepX Neo in Shanghai on July 13, a smartphone built around its own Step AOS operating system and an AI agent called Amoo that handles tasks across apps without user input. The launch puts a three-year-old Chinese startup ahead of Apple and OpenAI in shipping agent-native hardware, with a fuller debut set for the World Artificial Intelligence Conference on July 17.

Multi-Agent Systems Need More Than a Prompt

Most multi-agent demos look simple. An agent receives a request, delegates a few tasks, calls some tools, and returns an answer. Then you try to use the same system for real work. Suddenly, the important questions are not about the prompt: What context did each agent receive? Which workflow or tool ran? Where is the output from the previous step? Why did the process stop? Is the system waiting for a person? Can we continue without starting again? How much time and money did each step consume? If the answers are scattered across chat sessions, terminal logs, and database records, the system becomes difficult to trust. This is the problem we are working on with Heym. In a traditional Kanban board, cards describe work that people need to complete. In Heym's Agentic Kanban Board, a card is the work. Each column can contain an ordered chain of AI workflows. Moving a card into a column starts that chain automatically. Moving a card from Backlog to Planning is no longer just a status change. It is an execution event. The Planning column might: Analyze the request. Retrieve relevant context. Ask clarifying questions. Produce an implementation plan. Wait for human feedback. The Development column can then receive the same card, including the plan and the human's answers, and start a different workflow chain. The card becomes the handoff between agents, workflows, models, and people. One of the hardest parts of building multi-agent systems is preserving context across steps. A Heym card carries: The original request Comments and human responses File attachments Previous workflow outputs Movement and activity history Execution results Error states Every new workflow receives the accumulated context. This means you do not need to copy the same information into another prompt or invent a custom protocol for every agent handoff. The card acts as shared memory for the process. Different columns can use different models, tools, and workflows while operating on the same persistent j

ServiceNow: AI Is An Opportunity, Not A Threat

How we built Engrava: from cognitive-architecture research to a production library

Deterministic consolidation, a typed graph in SQLite, and an honest look at what agent-memory benchmarks can and can't measure. You're building an agent. It answers questions across many sessions, and by session three it has forgotten what it learned in session one. The reflex is to put a vector database in front of it - and now it remembers a blurry average of everything, ranked by cosine distance, contradicting itself and unable to tell a three-week-old preference from a stale throwaway. A graph database gives you structure, at the cost of a second persistence model with its own query language and deployment. A managed memory service starts you in a few lines, and moves the decision of what your agent remembers onto someone else's infrastructure. Underneath all of them is one problem: if every plausible fact is written the moment it appears, memory turns into an accumulation layer instead of a judgement layer. Engrava is our answer to that - local, structured, and deliberate about what it keeps in reach. Here is how it is built, and why. Before the storage design there was a long stretch of reading cognitive-architecture research, circling one question: what does a long-running agent actually need from memory if that memory has to stay inspectable? That framing is the reason Engrava is a typed graph and not a bag of embeddings, the reason consolidation is deterministic instead of an LLM rewrite pass, and the reason extraction stays above the database instead of hiding inside it. Engrava is a typed knowledge graph with hybrid search, in a single SQLite file. Thoughts are nodes; typed edges carry the relationships a flat vector can't - that A caused B, that C specializes D. Retrieval fuses vector similarity, keyword match, recency, and priority rather than leaning on cosine distance alone. Turn on the journal and every thought and edge mutation is recorded in a tamper-evident SHA-256 chain. It is all in the free package. Every thought has a priority, recomputed each

AI’s Biggest Winners Are Creating Its Biggest Losers. Here’s What It Means for Investors

What Is GPT? A Practical Guide to Tokens, Transformers, Training, and Fine-Tuning

Artificial intelligence systems can now write articles, explain scientific concepts, generate software code, summarize documents, and participate in remarkably natural conversations. At the center of this development is a class of language models commonly associated with three letters: GPT. Despite its widespread use, GPT is often described too simply. It is not merely a chatbot, a search engine, or a database containing prepared answers. GPT is a neural language model trained to process sequences of tokens and predict what should come next. Understanding GPT therefore requires looking beyond the chat interface. We need to examine tokenization, Transformer architecture, pre-training, parameters, post-training, and the statistical process through which a model produces language. What Does GPT Stand For? GPT stands for Generative Pre-trained Transformer. Each word describes a fundamental part of the system. Generative means that the model can produce new sequences, such as text, code, structured data, or other token-based outputs. Pre-trained means that the model first learns general patterns from a large collection of data before it is adapted for specific tasks or conversational behavior. Transformer refers to the neural-network architecture on which GPT is based. The Transformer architecture was introduced by Vaswani and colleagues in the 2017 paper Attention Is All You Need. Unlike earlier sequence models that depended heavily on recurrent neural networks, the Transformer used attention mechanisms to process relationships between elements in a sequence more efficiently and in parallel. The original GPT research applied generative pre-training to a Transformer-based language model. The central idea was to first train a general-purpose model on unlabelled text and then adapt it to downstream language tasks. This combination of large-scale pre-training and task-specific adaptation became one of the foundations of modern natural language processing. GPT Does Not Read