COGNOSCERE Daily Tech Review — Issue T145 · Friday, June 12, 2026

Or visit Intelligence Overview for deeper analysis.

ACTIvanti Tells Sentry Customers to Patch Now as Critical Bugs Hit 10.0 and 9.9

Ivanti disclosed two critical vulnerabilities in Sentry—CVE-2026-10520 (CVSS 10.0, unauthenticated root-level RCE via OS command injection) and CVE-2026-10523 (CVSS 9.9, authentication bypass enabling rogue admin account creation)—and released patches in versions 10.5.2, 10.6.2, and 10.7.1. Although Ivanti initially reported no in-the-wild exploitation, the Shadowserver Foundation subsequently observed active exploitation attempts based on a public PoC, with at least two exposed gateways already backdoored.

The Register · Cybersecurity · Relevance: 0.9 · Source →

remote-code-execution, authentication-bypass, patch-now, enterprise-security, mobile-gateway, zero-day

ACTHades PyPI Attack: 19 Packages Poisoned to Auto-Run Bun Credential Stealer

Attackers in the Hades campaign—an evolution of the Miasma/Shai-Hulud supply-chain worm lineage—compromised 37 malicious wheel artifacts across 19 PyPI packages by exploiting Python .pth startup hooks to auto-execute a Bun-powered JavaScript credential stealer at interpreter startup, harvesting cloud tokens (AWS, GCP, Azure), GitHub keys, SSH keys, and AI tool configurations. The exfiltrated data was sent to attacker-controlled GitHub repositories; organizations using affected packages should immediately remove them and rotate all reachable credentials.

The Hacker News · Cybersecurity / Supply Chain · Relevance: 0.9 · Source →

supply-chain-attack, credential-theft, Python, open-source, CI/CD, cloud-security, developer-tools

PREPAREBroadcom, Apollo, and Blackstone Establish Landmark Strategic Platform to Accelerate More Than 20 Gigawatts of Global AI Deployments

Broadcom launched the AI XPV Platform with Apollo and Blackstone as anchor investors, raising an initial $35 billion tranche to deliver more than 20 GW of AI compute capacity through 2028 using Broadcom's custom XPUs and networking solutions customized for frontier AI labs including Anthropic and OpenAI. The first tranche will directly fund Anthropic's previously announced expansion of more than 1 GW of compute infrastructure expected to deploy at Fluidstack-based sites starting mid-2026.

Apollo Global Management · Cloud / AI Infrastructure · Relevance: 0.9 · Source →

AI-infrastructure, data-center, compute, private-equity, semiconductor, XPU, AI-investment

WATCHSpaceX Details Its AI1 Orbital Data Center Satellite Ahead of IPO

SpaceX unveiled its AI1 satellite on June 8—a 70-meter wingspan, solar-powered orbital AI compute platform delivering 120 kW average and 150 kW peak compute capacity with an interchangeable chip payload—intended as the first generation of a planned constellation of up to one million AI satellites, timed to the company's historic IPO pricing on June 11. Anchor customers Anthropic and Google are paying for compute capacity, and SpaceX has announced an 11-million-square-foot Gigasat factory in Bastrop, Texas, to manufacture the satellites and solar components at scale.

Tom's Hardware · Data Infrastructure / Edge Computing · Relevance: 0.8 · Source →

AI-infrastructure, orbital-compute, IPO, data-center, edge-computing, power, satellite

PREPAREPolicy on the AI Exponential

Anthropic CEO Dario Amodei published a sweeping policy essay calling for the US government to move from transparency-only rules to binding FAA-style regulation of frontier AI, including mandatory third-party testing across four risk areas (cybersecurity, biological weapons, loss of AI control, and automated R&D) with legal authority to block or reverse model deployments. Alongside the essay, Anthropic released a legislative proposal on frontier model testing and pledged $350 million in new funding—a $200 million Economic Futures Research Fund and a $150 million national fellowship program—to address AI-driven labor displacement.

Dario Amodei (darioamodei.com) · AI Regulation & Governance · Relevance: 1.0 · Source →

AI-regulation, AI-safety, frontier-AI, policy, labor-displacement, governance, national-security

PREPAREDXC Will Integrate Claude Into the Systems Banks, Airlines, and Other Regulated Industries Rely On

Anthropic and DXC Technology announced a multi-year global alliance making DXC one of the few Global Premier partners in the Claude Partner Network; DXC will train tens of thousands of Claude-certified forward-deployed engineers to embed Claude into the mission-critical systems it operates for major banks, airlines, insurers, and government agencies across 70 countries. The partnership builds on DXC's existing OASIS platform—where Claude generates over 95% of code before human review and has already accelerated software delivery an estimated 10x—now deployed to more than 50 customers.

Anthropic · Enterprise AI Adoption · Relevance: 0.9 · Source →

enterprise-AI, managed-services, agentic-AI, IT-services, banking, insurance, government

WATCHGoogle DeepMind Is Worried About What Happens When Millions of Agents Start to Interact

Google DeepMind, in partnership with Schmidt Sciences, ARIA, the Cooperative AI Foundation, and Google.org, announced a $10 million funding initiative for external researchers to study the safety risks that emerge specifically from large-scale multi-agent AI systems interacting autonomously—including collusion, emergent behaviors, and novel cyberattack vectors—a category the initiative's leaders say has virtually no dedicated research field yet. Applications are open through August 8, with awards expected in autumn 2026.

MIT Technology Review · AI Safety & Alignment · Relevance: 0.9 · Source →

multi-agent-AI, AI-safety, agentic-systems, research-funding, AI-governance, emergent-behavior, coordination

PREPAREOpenAI Considers Drastic Price Cuts, Anticipating War for Users With Anthropic

OpenAI is weighing significant token price cuts across its API in anticipation of similar moves by Anthropic, whose Claude Code tool has surged among software engineers; both companies recently filed confidential IPO prospectuses and are burning tens of billions annually on compute. A potential price war would compress already-thin margins for both pre-IPO companies, raising questions about the sustainability of enterprise AI cost models.

CNBC · Strategic Planning / Market Intelligence · Relevance: 0.9 · Source →

pricing-strategy, AI-market, IPO, competitive-intelligence, enterprise-AI, token-economics, OpenAI, Anthropic