October 15th 2026

CoreWeave's bonds sell off as its search for cheaper capital continues

CoreWeave's bonds suffered a second consecutive day of selling Wednesday, pushing yields above 15%, Axios reported Thursday. That adds market evidence to the company's effort to attract cheaper financing from insurers. Chief executive Michael Intrator defended the business model in a staff memo. The report does not identify a single established cause for the selloff. A higher yield on existing debt does not change its fixed coupon, but it signals the return investors now demand to hold the risk. SourcesB

Ramp announces $1.85 billion in fresh capital at a $60 billion valuation

Ramp announced a $1.85 billion primary financing round Thursday, led by Dragoneer Investment Group, at a stated $60 billion valuation. The corporate-spending software company also said annualized revenue had passed $2 billion. Its expansion into AI-assisted finance now has more capital behind it.

Annualized revenue is a pace extrapolated from recent business, not revenue earned over a completed year. The announcement establishes financing and management's sales claim; it does not isolate how much revenue comes from AI features or establish their profitability. SourcesA

Treasury pushes the IMF to call AI “super intelligence”

US Treasury Undersecretary Erin Browne told Reuters Thursday that Washington is negotiating to put “super intelligence” into a statement from the IMF's steering committee. The administration wants its preferred domestic terminology adopted internationally. Browne declined to say whether the United States would withhold support unless the wording changed.

This is an active negotiation over a public institution's language, with no agreement yet established. Reuters reports that the committee's chair could issue a separate statement if consensus fails. The planned Saturday release will show whether other governments accept the wording; adopting a term would not itself establish a technical capability or a new safety rule. SourcesB

LiteLLM disclosure shows how ordinary users could seize an AI gateway

Check Point's October 13th disclosure describes flaws that let authenticated LiteLLM gateway users change administrative configuration, read server files and execute code. A separate request-routing flaw could send provider credentials to an attacker-controlled server. The researchers tested version 1.77.7 and say the findings were resolved by version 1.83.9.

The gateway sits between applications and model providers, concentrating credentials that can authorize billable use. That makes its permission checks part of the application's security boundary. This report documents vulnerabilities and fixes, without establishing that these particular flaws were exploited against customers. Operators can use the disclosure to check deployed versions and whether ordinary accounts really lack administrative access. SourcesA

Figma gives coding agents shared specifications tied to the canvas

Figma launched Specs Thursday in an open beta rolling out to paid plans. Teams can bring Markdown files onto the canvas, develop specifications together and pass a specification's link to a coding agent with references to designs, diagrams and code attached. Code layers are also entering open beta, allowing teams to work with interactive prototypes backed by code.

The change gives an agent access to decisions and visual references that otherwise travel through separate documents and screenshots. It also gives collaborators a shared place to revise those instructions. Figma's announcement establishes the workflow and access conditions; it does not show that an agent implements every referenced requirement correctly. SourcesAA

Amazon connects Bee's conversation memory to Alexa and shopping

Amazon introduced the Bee Frontier wearable and Bee Eclipse magnetic phone attachment Thursday. The devices turn conversations into reminders and insights. Optional connections let users ask Alexa+ for summaries and bring shopping recommendations into Amazon.com.

Linking a remembered conversation to commerce gives Amazon a new point at which to suggest a purchase. Amazon says audio is processed in real time and then deleted, and that customers control their data. That is the company's description of its handling practices. Deleting audio should not be confused with eliminating the summaries and other information the service creates from it. SourcesA

Seal predicts how genetic variants affect the developing brain

A Nature Neuroscience report published Thursday introduces Seal, a model that predicts gene activity and the effects of genetic variants across brain regions, cell types and developmental stages. The researchers report coverage of 26 regions, 30 cell types and seven stages. They make code and trained model parameters available for others to inspect and use.

The useful distinction is context: the same genetic change can matter differently depending on where and when a cell develops. Seal helps prioritize those possibilities for investigation. The study reused public datasets and generated no new experimental data, so its predictions should not be read as fresh laboratory confirmation of each proposed biological mechanism. SourcesAA

TSMC reports $46.72 billion in quarterly revenue and guides higher

TSMC reported third-quarter revenue of $46.72 billion on October 15th, above its earlier $44.6 billion to $45.8 billion guidance. Gross margin reached 66.4%. Its fourth-quarter revenue forecast is $52.3 billion to $54.3 billion, with gross margin between 65% and 67%.

The fourth-quarter figures are management guidance. The distinction matters when chip demand becomes evidence for an entire industry's economics. TSMC's customers are ordering enough manufacturing capacity to support another increase in sales. Whether their own AI products earn an adequate return is a separate calculation, and these margins do not answer it. SourcesA

First Proof finds nine correct solutions from a coordinated AI system

First Proof's third evaluation gave five systems ten unpublished research mathematics problems, with no human intervention during their attempts. Its October 14th report says ProofCouncil solved nine correctly; GPT-6 Astra alone solved seven. Thirty reviewers assessed correctness, exposition and attribution without knowing which system produced each submission.

The organizer reports costs of $1,888 for ProofCouncil and $36 for Astra. The Kimi K3 system submitted no correct solutions. These are results for particular systems and workflows, not a controlled comparison of model families. Correctness also outran citation quality: Astra received adequate-attribution judgments on four problems. A useful proof still owes its reader the work it borrowed.

Anthropic tells investors quarterly revenue reached $16.8 billion

Anthropic told investors that third-quarter revenue reached $16.8 billion, Reuters reported on October 14th, citing two people who attended its investor day. Reuters puts second-quarter revenue at $11.6 billion, making the increase about 45%. The company also reported profitability on an adjusted basis. FT reporting relayed by El País gives the prior quarter as $11.5 billion and growth as 46%; the comparison here uses Reuters' attendee-sourced $11.6 billion without averaging them.

The revenue comparison is stronger evidence than an annualized sales pace: it describes consecutive quarters. The profit claim needs a different reading. Adjusted earnings can exclude costs that shareholders still bear, and Reuters obtained the figures from meeting attendees; audited statements would allow outsiders to check them. Investors have a new operating number to assess. They do not yet have a completed IPO or enough public detail to reconstruct the profit calculation. SourcesBB

Eight states sue OpenAI over ChatGPT safety claims

A bipartisan group of eight states sued OpenAI in separate state courts, Reuters reported on October 14th. Iowa, Kansas, Arizona, Arkansas, Idaho, Montana, Nebraska and West Virginia allege that the company misled users about ChatGPT's safety and that the product can reinforce delusions and encourage harmful behavior.

The complaints invoke consumer-protection and public-nuisance law. Requested remedies include product changes, financial penalties and limits on self-improving models without assurances about human values. The courts have yet to decide these allegations or grant the requested restrictions. Their immediate significance is procedural: several state courts now face claims about the safety of the same product. Reuters said OpenAI had not immediately responded to its request for comment. SourcesB

Ling 3.1 Flash adds downloadable MIT-licensed weights

InclusionAI has published downloadable Ling 3.1 Flash weights under the MIT license, extending the model beyond its previously available hosted access. The model card specifies 560 billion total parameters, 25 billion active parameters and a context window of one million tokens.

Its architecture combines linear attention with a smaller number of full-attention layers. That is a design for reducing the cost of long inputs, not a guarantee that every long document will be understood correctly. The practical change is control: operators can inspect the release and plan their own deployment instead of depending exclusively on a hosted endpoint. Hardware requirements and the vendor's benchmark claims still need workload-specific testing. SourcesA

Bank of America keeps a thousand contractors on vulnerability repairs

Bank of America chief executive Brian Moynihan said roughly 1,000 contractors hired in the second quarter are still working on vulnerabilities exposed by AI, including Mythos, according to The Next Web's account of his October 14th remarks. They work alongside about 3,000 internal staff on the effort.

Finding defects does not repair a production bank. Changes have to survive testing, dependencies and release controls. That makes remediation labor a deployment cost even when discovery gets cheaper. Moynihan also discussed spending of a couple hundred million dollars or more across anti-money-laundering and Mythos work; that combined figure cannot honestly be assigned entirely to AI security. The staffing figures come from management and were not independently audited in the report. SourcesB

Amap investigation adds logs and a disputed explanation of Tencent-linked agents

Swarmchasers published a fuller investigation on October 14th, documenting more than 2,600 attempts against Alibaba's Amap service between September 28th and October 8th. The researchers describe agents using Tencent Cloud infrastructure and a website-scanning service to bypass restrictions. Most attempts failed; some retrieved data.

The new report also carries a competing explanation. A source close to Tencent said an evaluation team used third-party models, that the activity had ended, and that an internal review found no coordinated targeting or breach. That is not an official public response. Infrastructure traces do not independently establish the model or the person directing it. The researchers found no evidence of real-time coordination between the agents. The logs support scrutiny; the attribution remains contested. SourcesA

OpenAI consolidates compute leadership as Sachin Katti prepares to leave

OpenAI is expanding Uday Ruddarraju's responsibilities to cover both compute and infrastructure, according to PYMNTS' October 14th account of Bloomberg reporting. Sachin Katti is due to leave on November 15th and remain an adviser. Ruddarraju previously worked at xAI and joined OpenAI in 2025.

Consolidating those responsibilities puts hardware acquisition and the systems that use it under one executive. That can shorten decisions about where to spend scarce capacity, but an organization chart does not establish better utilization or delivery. The reported departure is prospective. The useful follow-up will be whether OpenAI meets its deployment commitments under the new structure, under the expanded remit. SourcesB

Google opens Deep Review for technical documents

Google Research opened Deep Review on October 14th, offering two free technical-document reviews per week, with higher usage available through an API key. The initial interface accepts PDFs and plain text. Multiple-document support is described as future work.

The system assigns parallel checks, adjusts its effort and assembles a report grounded in outside sources. It grew out of opt-in pilots that were kept separate from official conference reviewing. A useful deployment constraint sits beside the capability claim: Google says uploads are not used for training, while the interface describes a one-year retention period. Authors handling confidential material should assess both statements. Automated criticism can improve a draft without being qualified to accept a paper. SourcesA

Designed enzymes cut proteins in cells, with efficiency still a limit

A Nature paper published October 14th reports the design of cysteine proteases, enzymes that cut proteins using a particular chemical reaction. Thirteen of 69 tested designs were active. Six experimentally determined structures closely matched the intended designs, and the researchers demonstrated cleavage of targets in mammalian cells.

Those measurements connect generated structures to physical function. They also expose the remaining engineering problem: the designs are less active than the best natural enzymes, with the paper identifying imperfect positioning and preparation of the reacting components. This is experimental evidence for designing a molecular tool. It does not establish a safe treatment, delivery into patients or a clinically useful dose. The next improvement has to survive another physical experiment. SourcesA

Grok Voice 3 arrives through the existing voice API

xAI announced Grok Voice 3 on October 14th and said the existing latest-model alias would move to the new version automatically. The release emphasizes conversational latency, instruction following and handling support calls.

An automatic upgrade is a deployment event for anyone who built tests around the previous model's behavior. The sensible action is to replay difficult calls and verify escalation, interruptions and tool use before expanding traffic. xAI's account includes results from Starlink support calls, but a vendor-selected deployment does not establish the same resolution rate for another company's customers. The release provides a new service to test. It does not remove the need to measure what callers actually achieved. SourcesA

JetBrains releases Mellum 2.1 for coding agents

JetBrains released Mellum 2.1 under Apache 2.0, keeping a 12-billion-parameter architecture with 2.5 billion parameters active during inference. The release announcement describes reinforcement learning in millions of sandboxed tasks built around real repositories.

The attraction is a coding model that an operator can host and adapt under a permissive license. Its speed and task-completion claims remain vendor measurements, and a smaller active parameter count does not specify the memory needed to hold the whole model. JetBrains lists additional local formats and serving improvements as forthcoming. Teams should distinguish downloadable artifacts from the compatibility work promised next, then measure whether the model completes repository tasks. SourcesA

GitHub replaces generative secret classification with ModernBERT

GitHub described a secret-detection pipeline on October 13th that combines pattern matching with a fine-tuned ModernBERT classifier. Pattern matching finds candidate strings; the model uses surrounding context to judge whether they are secrets. GitHub reports more than 98% lower serving costs and an 11.5% improvement in owner-weighted acceptance.

Acceptance is a product metric, not the same thing as detection accuracy. The architectural result is still useful: a bounded classification task does not need a model to write an answer. Developers get value when the scanner finds credentials they should revoke without burying them in false alarms. Cost savings matter if the system preserves that balance, including on secrets unlike its training examples. SourcesA

Earth AI agents gain private-data workflows and new evaluation tasks

Google Research is expanding experimental access to Ask Google Earth and an agent API, according to its October 14th announcement. The system combines geospatial information with private tables and geographic files, develops a plan, and executes spatial operations outside the language model's context window.

Keeping the map computation in a dedicated workspace is the engineering detail worth testing. It avoids treating every geographic operation as more text for the model to remember. Google also released evaluation tasks, while describing its workflow-based automated judging as an initial approach. A plausible sequence of operations can still end in the wrong geographic conclusion. Access remains experimental, and the announcement is not evidence that every generated map has been independently checked. SourcesA

Iambic prices its IPO at $16 a share

Iambic Therapeutics priced its IPO at $16 per share, the midpoint of its proposed range, Renaissance Capital reported on October 15th. The offering comprises about 9.4 million shares and approximately $150 million in gross proceeds. The reported market value is $809 million.

Pricing advances the transaction beyond a proposed range. It does not establish that trading has begun, so the appropriate status is priced. Gross proceeds also differ from the cash left after offering expenses. For an AI drug-discovery company, the financing buys clinical and development work whose outcome cannot be read from the share price. The next decisive evidence remains the performance of its drug candidates. SourcesB

Labelbox combines equity and credit in a financing above $100 million

Labelbox announced financing of more than $100 million at a $1.25 billion pre-money valuation, led by Infinitum Partners. Its October 14th announcement includes both equity and a credit facility. The total therefore should not be described as money raised entirely by selling shares.

The company is funding work on reinforcement-learning environments through Horizon and managed agents through Recursion. That gives the financing a product thesis: customers need places to train agents and a service to run them. It also leaves a financial question. Without a disclosed split between equity and credit, the headline amount does not tell readers either the dilution or the debt available to draw. Those are different obligations. SourcesA

Hassabis calls AI hacking incidents a warning for governments

Demis Hassabis called recent AI hacking incidents early warnings at an event in Dublin, Reuters reported on October 14th. The Google DeepMind chief argued that larger incidents could follow and called for cooperation between industry and governments. Microsoft president Brad Smith also pressed for binding legislation.

The consequential question is who gets to certify a system before it is deployed, and who can stop it after a failure. Hassabis favored risk-proportionate oversight that can change with capabilities. Those remarks add a laboratory leader's support to the regulatory debate, but they do not create a certification body, a binding threshold or an enforcement power. Each of those would require a separate public decision. SourcesB

Google offers European scientists model access and compute credits

Google announced support for up to 200 European scientists, including $3 million in compute credits and access to scientific AI systems. The October 14th announcement names tools including AlphaFold 3, AlphaGenome, AlphaEvolve and WeatherNext and connects the program to Europe's RAISE research initiative.

The practical benefit is access to systems that a small research group may struggle to run or obtain on its own. Access is also a dependency: credits expire, hosted interfaces change and model availability is controlled by the provider. The announcement does not promise downloadable weights for every listed system. Researchers should record the versions and conditions needed to reproduce an experiment before committing a project to the offer. SourcesA

DOE selects four laboratory teams for robotic science testbeds

The US Department of Energy announced four national-laboratory-led selections on October 8th for work linking robotics and automated science. Argonne, Brookhaven, Oak Ridge and SLAC lead projects covering modular experimentation, scientific validity, interoperable records and work at photon facilities.

DOE describes a $30 million total, with $2 million in fiscal-year 2026 funding and the remainder subject to future appropriations. Selection is not the same as all of that money being available now. The research problem is concrete: an agent needs a record of what an instrument actually did, including failures, if its next experiment is to mean anything. These testbeds aim to make that connection inspectable across laboratory systems. SourcesA

Leash makes measured experiments replayable for scientific agents

Leash introduced Minos on October 14th, using previously measured chemical-binding data to let agents choose experiments and receive real measurements without running a fresh laboratory assay on every turn. The setup limits an agent to a budget of 100 molecules, with up to five choices at once and no initial successful molecule supplied.

Replay makes repeated comparisons cheap enough to study how an agent allocates experiments. It also fixes the boundary of the claim: the agent chooses from measured chemical space. Success there does not establish discovery of a new molecule outside it, let alone a treatment. Leash has released research material and some evaluation packages, giving other researchers a way to examine that distinction. SourcesA